JP6435803B2 - Order quantity determination device, order quantity determination method and order quantity determination program - Google Patents

Description

  The present invention relates to an order quantity determination device, an order quantity determination method, and an order quantity determination program.

  There is a technique for predicting a demand amount of a product and determining a safe order quantity that does not cause a stock out of the product with the predicted demand amount. In such a technique, a safety stock quantity is calculated by multiplying a predicted demand amount of a product by a safety coefficient, and an order quantity is determined using the safety stock quantity.

International Publication No. 2004/022463

  By the way, an ordering person in charge of ordering at a retail store is required to perform an optimal ordering operation. The meaning of “optimal” varies depending on the situation and the circumstances of the ordering party. For example, retail stores have a desire not to run out of stock. In addition, retail stores have a desire to increase profits. In addition, there is a demand for retailers to level out the order quantity in order to reduce the work burden when delivering the ordered product and to reduce the burden on the producer of the ordered product.

  However, with the conventional technology, there are cases where it is not possible to place an order according to the actual situation of the ordering work. For example, in the conventional technique, since the order quantity is determined using the safety stock quantity, the order quantity varies greatly, and there are cases where a large number of products are delivered at once and the work load is heavy. Further, in the conventional technology, since the number of delivered products is greatly varied, the variation in the work amount at the time of delivering the ordered products is also large, and it is difficult to shift the workers in the retail store.

  In one aspect, an object of the present invention is to provide an order quantity determination device, an order quantity determination method, and an order quantity determination program that can determine an order quantity according to the actual situation of ordering work.

  In the first plan, the order quantity determination device includes a reception unit and a calculation unit. A reception part receives the demand forecast value of goods. The calculation unit calculates the order quantity of the product that minimizes the sum of the index value related to the change in the order quantity of the product and the index value related to the order cost of the product based on the received demand prediction value.

  According to one embodiment of the present invention, there is an effect that it is possible to determine the order quantity according to the actual situation of the ordering work.

FIG. 1 is a functional block diagram illustrating the configuration of the order quantity determination device according to the first embodiment. FIG. 2 is a diagram illustrating an example of the condition designation screen. FIG. 3 is a diagram showing the relationship between continuous time and discrete time related to ordering. FIG. 4 is a diagram schematically illustrating demand prediction. FIG. 5A is a diagram illustrating an example of a demand prediction result. FIG. 5B is a diagram illustrating an example of a demand prediction result for each component. FIG. 6 is a diagram illustrating an example of an output screen. FIG. 7A is a diagram showing changes in the order quantity and the completely leveled order quantity according to the conventional safety stock quantity. FIG. 7B is a diagram illustrating a change in the inventory amount. FIG. 8A is a diagram illustrating a change in the order quantity when the order quantity and the order cost based on the conventional safety stock quantity are most emphasized. FIG. 8B is a diagram illustrating a change in the inventory amount. FIG. 9A is a diagram illustrating an example of an output screen. FIG. 9B is a diagram illustrating a change in the inventory amount. FIG. 10 is a flowchart illustrating an example of the order quantity determination process. FIG. 11 is a diagram illustrating a computer that executes an order quantity determination program.

  Embodiments of an order quantity determination device, an order quantity determination method, and an order quantity determination program disclosed in the present application will be described below in detail with reference to the drawings. This scope of rights is not limited by this embodiment. Each embodiment can be appropriately combined within a range in which processing contents do not contradict each other.

  An example of the overall configuration of the order quantity determination apparatus 10 according to the first embodiment will be described. FIG. 1 is a functional block diagram illustrating the configuration of the order quantity determination device according to the first embodiment. The order quantity determination device 10 is an apparatus that determines the order quantity of a product. The order quantity determination device 10 predicts demand for a predetermined period for a product to be ordered, and obtains and outputs an order quantity according to the actual situation based on the predicted demand. The order quantity determination device 10 is, for example, a computer such as a personal computer or a server computer. The order quantity determination device 10 may be implemented as a single computer or may be implemented by a plurality of computers. In the present embodiment, the case where the order quantity determination device 10 is a single computer will be described as an example.

  As illustrated in FIG. 1, the order quantity determination device 10 includes a communication I / F (interface) unit 20, an input unit 21, a display unit 22, a storage unit 23, and a control unit 24. Note that the order quantity determination device 10 may include devices other than the above devices.

  The communication I / F unit 20 is an interface that controls communication with other devices. As the communication I / F unit 20, a network interface card such as a LAN card can be adopted.

  The communication I / F unit 20 transmits and receives various types of information to and from other devices via a network (not shown). For example, the communication I / F unit 20 can transmit and receive various types of information to and from a system that manages the ordering and receiving of products, and transmits and receives various types of information related to products to be ordered from the system.

  The input unit 21 is an input device that inputs various types of information. Examples of the input unit 21 include an input device that receives an input of an operation such as a mouse or a keyboard. The input unit 21 receives input of various types of information. For example, the input unit 21 receives various operation inputs related to determination of the order quantity. The input unit 21 receives an operation input from the user, and inputs operation information indicating the received operation content to the control unit 24.

  The display unit 22 is a display device that displays various types of information. Examples of the display unit 22 include display devices such as an LCD (Liquid Crystal Display) and a CRT (Cathode Ray Tube). The display unit 22 displays various information. For example, the display unit 22 displays various screens such as various conditions relating to ordering and a screen for displaying the determined order quantity. For example, the display unit 22 displays a condition designation screen described later, a screen displaying the determined order quantity, and the like.

  The storage unit 23 is a storage device such as a hard disk, an SSD (Solid State Drive), or an optical disk. The storage unit 23 may be a semiconductor memory that can rewrite data, such as a random access memory (RAM), a flash memory, and a non-volatile static random access memory (NVSRAM).

  The storage unit 23 stores an OS (Operating System) executed by the control unit 24 and various programs. For example, the storage unit 23 stores various programs used for determining the order quantity. Furthermore, the storage unit 23 stores various data used in programs executed by the control unit 24. For example, the storage unit 23 stores product information 30, demand record information 31, demand prediction information 32, and order quantity data 33.

  The product information 30 is data that stores various types of information related to the product to be ordered. The product information 30 stores various types of information used for determining the order quantity, such as the inventory quantity of the product to be ordered, the order unit, and the unit price of the order.

  The demand record information 31 is data in which information related to past demand related to a product to be ordered is stored. For example, the past demand amount of the product to be ordered is stored in the demand record information 31.

  The demand prediction information 32 is data that stores information related to a predicted demand related to a product to be ordered. For example, the demand prediction information 32 stores a predicted demand amount for a product to be ordered.

  The order quantity data 33 is data that stores information related to ordering of products to be ordered. For example, the order quantity data 33 stores the order quantity calculated for the product to be ordered.

  The control unit 24 is a device that controls the order quantity determination device 10. As the control unit 24, an electronic circuit such as a CPU (Central Processing Unit) and an MPU (Micro Processing Unit), or an integrated circuit such as an ASIC (Application Specific Integrated Circuit) and an FPGA (Field Programmable Gate Array) can be employed. The control unit 24 has an internal memory for storing programs defining various processing procedures and control data, and executes various processes using these. The control unit 24 functions as various processing units by operating various programs. For example, the control unit 24 includes a collection unit 40, a reception unit 41, a prediction unit 42, a calculation unit 43, and an output unit 44.

  The collection unit 40 performs various collections. For example, the collection unit 40 collects various types of information related to the product to be ordered. For example, the collection unit 40 collects the inventory amount of the product to be ordered, the unit of ordering, the unit price of ordering, and the like from the system that manages the ordering of products. The collection unit 40 causes the product information 30 to store various pieces of information related to the collected products to be ordered. Further, the collection unit 40 collects the past demand amount of the order target product from the system for managing the order of the product, and stores the past demand amount of the order target product in the demand record information 31. In the present embodiment, the product information 30 and the demand performance information 31 are collected and stored by the collection unit 40 from a system that manages the ordering of products. However, the present invention is not limited to this. The product information 30 and the demand record information 31 may be stored by another system or an administrator.

  The reception unit 41 receives various conditions regarding ordering. For example, the reception unit 41 receives designation of what is important with respect to ordering. For example, the accepting unit 41 accepts designation of whether ordering leveling and ordering cost are important. For example, the receiving unit 41 displays a later-described condition specifying screen on the display unit 22, and receives from the condition specifying screen specification of a ratio indicating whether ordering leveling and ordering cost are important.

  FIG. 2 is a diagram illustrating an example of the condition designation screen. The condition designation screen 50 is provided with a track bar 51 for designating which one of leveling of order quantity and order cost should be emphasized. The track bar 51 is slidably movable in one and the other. When the track bar 51 is moved to one side, the order quantity is emphasized. When the track bar 51 is moved to the other side, the order cost is regarded as important. The orderer operates the track bar 51 on the condition designation screen 50 to designate which of the order quantity leveling and the ordering cost is important. Thereby, the order quantity determination device 10 calculates the order quantity of the product according to the designation and determines the order plan.

  Returning to FIG. 1, the prediction unit 42 performs processing related to the demand for goods. For example, various predictions regarding the demand for goods are performed. For example, the prediction unit 42 calculates the demand of the order target product for a predetermined period for obtaining the order quantity using a predetermined prediction model based on the past demand history of the order target product stored in the demand record information 31. Predict. Generally, demand fluctuates periodically in units such as weeks, months, and years. For this reason, as a prediction model, the prediction model corresponding to a periodic fluctuation is used. For example, the prediction unit 42 predicts the upper limit demand generated with a predetermined probability using a linear Gaussian state space model. In addition, a prediction model is not limited to this, Another prediction model may be sufficient. Moreover, although a present Example demonstrates the case where the prediction part 42 estimates the demand of goods, it is not limited to this. The prediction unit 42 may not only perform the demand prediction itself, but also accept a demand prediction value calculated by another information processing apparatus. That is, the reception unit 41 may receive a demand forecast value for a product from another information processing apparatus.

  Here, variables and expressions used in the following description will be described. For example, when ordering periodically at a predetermined ordering interval, the timing of each ordering is shown as in FIG. FIG. 3 is a diagram showing the relationship between continuous time and discrete time related to ordering.

Here, t represents a continuous time. and t∈R ^1. R ¹ represents a one-dimensional real number. k is a discrete time indicating the ordering time. Let kεN ¹ . N ¹ represents a one-dimensional natural number. h represents an ordering interval. and h∈R ^1.

In addition, the inventory quantity of the product to be ordered at time kh is represented as x [k]. Let x [k] εZ ¹ ₊ . Z ¹ ₊ represents a one-dimensional positive integer. The actual amount of inventory is expressed as x _r [k] by adding “r” indicating actuality with a subscript. The predicted inventory quantity is expressed as x _p [k] with “p” indicating the prediction appended with a subscript.

Further, the order quantity of the product to be ordered at time kh is represented as u [k]. Let u [k] εZ ¹ ₊ .

In addition, the demand amount of the product to be ordered at time kh ≦ t <(k + 1) h is represented as d [k]. Let d [k] εZ ¹ ₊ . The actual demand is expressed as d _r [k]. The predicted demand is expressed as d _p [k].

  Here, in this embodiment, in order to simplify the explanation, it is assumed that the order quantity at the previous order time is delivered as it is. That is, the product with the order quantity u [k] at time kh is delivered at time (k + 1) h. In this case, the relationship of the following formula (1) is established for the inventory amount of the product. This formula (1) is calculated by adding the order quantity of the previous order time to the inventory quantity of the previous order time and subtracting the demand amount from the previous order time. Value.

Σ _x : x [k + 1] = x [k] + u [k] −d [k] (1)

Therefore, the Σ _x, is a dynamic system that stock amount of the current time is determined depending on the amount of inventory in the past of time.

  The formula (1) for calculating the stock quantity is an example, and the order quantity u [k] before the lead time is added in consideration of the lead time from when the order is placed until delivery. Also good.

  Next, prediction of demand using a linear Gaussian state space model will be described. In the linear Gaussian state space model, the next state is predicted from the previous state using the following equations (2) and (3).

ζ [k + 1] = A · ζ [k] + B · ν [k] (2)
d _p [k + 1] = C · ζ [k] + ω [k] (3)
k = 1, 2,...

Here, it is assumed that _＋ k∈N ₊ . N ₊ represents a positive integer.
In addition, the ζ [k] ∈R ^n. R ⁿ represents an n-dimensional real matrix. ζ [k] is ζ [k] to N (E [ζ [k]], P [k]).
Further, ν [k] εR ^r . R ^r represents an r-dimensional real matrix. ν [k] is ˜N (0, Q), which is a white Gaussian noise vector.
Also, ω [k] εR ^p . R ^p represents a p-dimensional real matrix. ω [k] is ˜N (0, R), which is a white Gaussian noise vector.
N (α, β) represents a normal distribution with mean α and variance β.
A, B, and C are execution columns having appropriate sizes. E [•] is an operation for calculating an average value.

  The orders of the states ζ, ν, and ω, and A, B, and C are determined in advance. For example, based on the past demand data, what order should be included from the trend of demand of the product to be ordered, season, AR (short-term circulation fluctuation component), MA (moving average component), event, weather, etc. Select arbitrarily. Then, the orders of the states ζ, ν, and ω and the sizes of A, B, and C are temporarily set, and E [ζ [1]], P [1] are obtained by the maximum likelihood method with the past demand data. , Q, R are determined. Then, for example, the accuracy is evaluated with the determined model by AIC (Akaike's Information Criterion), and when the evaluation does not satisfy the standard, the setting is repeated again.

  Here, for example, a prediction model having an l-order trend component can be expressed as the following equations (4) and (5).

  When the state is the following equation (6), the prediction model can be expressed as the following equations (7) and (8).

ζ [k] = [t [k], t [k−1],..., t [l]] ^T (6)
ζ [k + 1] = A _t · ζ [k] + B _t · ν _t [k] (7)
d _p [k] = C _t · ζ [k] + ω _t [k] (8)

  In addition, for example, a prediction model having a seasonal variation of 1 cycle can be expressed as the following equations (9) and (10).

  When the state is the following expression (11), the prediction model can be expressed as the following expressions (12) and (13).

ζ [k] = [s [k], s [k−1],..., s [k−l]] ^T (11)
ζ [k + 1] = A _s · ζ [k] (12)
d _p [k] = C _t · ζ [k] + ω [k] (13)

  Further, for example, when a prediction model having an l-th order trend component and an m-th order periodic variation has a state expressed by the following expression (14), the prediction model is expressed by the following expressions (15) and (16). I can express.

  In this embodiment, the following state is selected from the trend, season, and AR, and the next state is predicted from the previous state using, for example, the following equations (17) and (18) as the prediction model. .

  In this embodiment, the prediction is corrected by filtering based on the actual demand. For example, the prediction of the state one period ahead can be expressed as the following equations (19) and (20).

E [ζ _P [k]] = A · E [ζ _e [k−1]] (19)
P _P [k] = A · P _e [k−1] · A ^t + B · Q · B ^t (20)
Here, k = 1 is the initial value of the Kalman filter.
E [ζ _e [1]] = E · ζ [1]
P _e [1] = P [1]

For example, when the actual demand d _r [k] is obtained, the prediction is corrected using the following equations (21) to (23) by the Kalman filter.

K [k] = P _P [k] · C ^t (C · P _P [k] · C ^t + R) ⁻¹ (21)
E [ζ _e [k]] = E [ζ _P [k]] + K [k] (d _r [k]
-C · E [ζ _P [k]]) (22)
P _e [k] = P _p [k] −K [k] · C · P _p [k] (23)

  The demand as a result of the forecast is obtained as a probability distribution. For example, long-term demand ahead of i (i = 1,..., H) with the present k period can be expressed by the following equations (24) to (27).

  The predicting unit 42 predicts the upper limit demand generated one period ahead with a predetermined probability based on the past demand history of the product to be ordered stored in the demand record information 31. Then, the prediction unit 42 predicts the long-term demand by repeatedly predicting the upper limit demand generated by one period ahead with a predetermined probability, using the prediction result of forecasting one period ahead. In this embodiment, the upper limit demand occurring with a probability of 97.6% is predicted. FIG. 4 is a diagram schematically illustrating demand prediction. FIG. 4 shows a flow for sequentially predicting the upper limit demand generated with a probability of 97.6%. This prediction result is represented by a conditional probability.

  When the upper limit demand generated with a probability of 97.6% is predicted, the prediction of the state one period ahead can be expressed as the following equation (28). Further, the mean and variance of conditional probabilities can be expressed by the following equations (29) and (30).

  When the upper limit demand generated with a probability of 97.6% is predicted, the prediction of the state of i = 2... H period ahead can be expressed as the following equation (31). Further, the mean and variance of the conditional probability can be expressed as the following equations (32) and (33).

  The prediction unit 42 uses the above-described formulas (28) to (33) to calculate the upper limit demand generated with a probability of 97.6% from the past demand of the ordering target product stored in the demand performance information 31. Predict sequentially. The prediction unit 42 stores the prediction result in the demand prediction information 32. In addition, when the actual demand is obtained, the prediction unit 42 corrects the prediction parameters by the above-described Kalman filter based on the actual demand, and calculates the upper limit demand generated with a probability of 97.6%. The prediction is sequentially performed again, and the prediction result is stored in the demand prediction information 32.

  FIG. 5A is a diagram illustrating an example of a demand prediction result. FIG. 5A shows a result of forecasting demand while repeating correction using the Kalman filter based on the actual demand and the actual demand before the forecast time. The predicted demand is shown by a graph with an average and an upper limit where the variance is 2σ.

  FIG. 5B is a diagram illustrating an example of a demand prediction result for each component. FIG. 5B shows a graph in which the average predicted demand and the predicted demand are divided into trend, season, and AR components.

  The calculation unit 43 performs various calculations. For example, the calculation unit 43 calculates the order quantity of the product to be ordered based on the demand predicted by the prediction unit 42. For example, the calculation unit 43 solves the objective function optimization problem that minimizes the change in the order quantity of the product to be ordered and the order cost of the product according to the respective weights, thereby obtaining the order quantity of the product. Is calculated.

Here, the objective function and constraint conditions used for the optimization problem will be described. The order cost is determined, for example, as a function f _k (u [k]) of the order quantity u [k] of the product and the unit price of the product. For example, in the present embodiment, the ordering unit of the product to be ordered is one unit and a predetermined number of lot units, and single unit ordering and a predetermined number of lot orders can be performed. It is assumed that the unit price is practically different between lot order and lot order. For example, it is assumed that a single item and an order of 500 lots can be ordered for a product to be ordered, and the unit price is 460 yen. Further, as for the products to be ordered, when 500 lots are ordered, 50 products are discounted and the cost for 450 products is the ordering cost. The order cost of such goods in the k period can be expressed as the following equation (34).

f _k (u [k]) = floor (u [k] / 500) × 460 × 450
+ Rem (u [k], 500) × 460 (34)
Here, floor (a / b) is an operation for obtaining a value of an integer part obtained by dividing a by b. rem (a, b) is an operation for obtaining a remainder value obtained by dividing a by b.

  The formula (34) for calculating the ordering cost is an example, and is determined according to the ordering unit. The ordering cost may be calculated in consideration of other costs such as delivery costs.

  Therefore, the ordering cost from the k + 1 period to the H period ahead can be expressed by the following equation (35).

The J _c, the more the value increases many orders cost.

In addition, there may be an upper limit or a lower limit for the order quantity of merchandise depending on the relationship between the manufacturer and delivery. In this embodiment, the order target product is assumed that there is an upper limit U _u and a lower limit L _u as in the following (36) to the order quantity u [k].

L _u ≦ u [k] ≦ U _u (36)

  Further, when the order quantity is leveled, the fluctuation of the order quantity becomes small. The level of order quantity from the (k + 1) period to the H period ahead can be expressed as (37) below.

The J _e, change in order quantity is small, as has been leveled value increases. The equation (37) for calculating the leveling of the order quantity is calculated by summing up the difference from the order quantity one period before, but is not limited to this example. For example, the degree of order quantity leveling may be obtained by summing up the differences between the order quantities before a plurality of periods. For example, the level of order quantity is the difference between the order quantity for the k period and the order quantity for the k-1 period, and the difference between the order quantity for the k period and the order quantity for the k-2 period. You may calculate in total.

  In addition, when a product is out of stock, a sales opportunity is lost. For this reason, in order not to run out of stock for the product to be ordered, it is sufficient if there is a stock that is more than the demand predicted in each period. In this embodiment, the upper limit demand generated with a probability of 97.6% is sequentially predicted. For this reason, from the k + 1 period to the H period ahead, the condition that the stockout does not occur with a probability of 97.6% can be expressed as (38) below.

  In the present embodiment, the following equation (39) is used as an objective function from the equations (35) and (37).

Here, k is a weighting coefficient, and 0 ≦ k ≦ 1.
maxJ _C is the maximum order quantity that can be ordered from the k + 1 period to the H period.
maxJ _e is the maximum ordering cost can be ordered from the k + 1 phase to H-life destination.

For example, when the orderable range of products to be ordered is 0 to 10000, and U _u = 0 and L _u = 10000, maxJ _C is calculated from the k + 1 period as shown in the following equation (40). It is obtained by summing up the maximum order quantity that can be ordered up to H period ahead. As shown in the following formula (41), maxJ _e is obtained by summing up the maximum fluctuation range in the orderable range from the k + 1 period to the H period.

  In accordance with the track bar 51 on the condition designating screen 50, the calculation unit 43 increases the ordering cost with an emphasis on leveling the order quantity within the range of 0 to 1 with respect to the weighting factor k described above (39). The smaller the value is, the more important it is. Then, the calculation unit 43 sets the objective function of the equation (39) to be the minimum with the restriction of the order quantity shown in the equation (36) and the limitation of the inventory amount shown in the equation (38) as constraints. The order quantity from the (k + 1) period to the H period ahead is obtained by solving the optimization problem for obtaining the ordered quantity. For example, the calculation unit 43 obtains the value of Expression (39) by changing the order quantity for each period from a predetermined initial value to plus or minus. When the calculated value becomes smaller, the calculation unit 43 changes the order quantity of each period so that the value of Expression (39) becomes smaller, and the order quantity of each period when the value of Expression (39) does not decrease. Is determined as the optimal solution. The calculation unit 43 stores the calculated order quantity for each period in the order quantity data 33.

  The output unit 44 performs various outputs. For example, the output unit 44 displays a graph indicating the demand predicted by the prediction unit 42 and stored in the demand prediction information 32 and the order quantity calculated by the calculation unit 43 and stored in the order quantity data 33. Output to.

  FIG. 6 is a diagram illustrating an example of an output screen. In the example of FIG. 6, two graphs are displayed that are displayed according to the degree of importance of order quantity leveling and order cost. For example, the graph on the left side of FIG. 6 shows the predicted demand and the order quantity when the importance is placed on leveling the order quantity. In the graph on the left side of FIG. 6, the order quantity is most important, so the order quantity is indicated by a straight line near 5000 pieces. The graph on the right side of FIG. 6 shows the predicted demand and the order quantity when the order cost is most important. In the graph on the right side of FIG. 6, since the ordering cost is most important, the order quantity changes according to demand. The output unit 44 displays a waveform graph between the graph on the left side and the graph on the right side in FIG. 6 at a weighted ratio according to the designation of the track bar 51 on the condition designation screen 50.

  Here, the relationship between the order quantity and the inventory quantity will be described. FIG. 7A is a diagram showing changes in the order quantity and the completely leveled order quantity according to the conventional safety stock quantity. In the example of FIG. 7A, the change in the order quantity calculated from the safety stock quantity is indicated by a broken line in accordance with the predicted demand. Further, in the example of FIG. 7A, the order quantity that is completely leveled according to the predicted demand is shown by a straight line in the vicinity of 5000 pieces. In the example of FIG. 7A, the ordering cost based on the safety stock amount is 155,234,360 yen. Completely leveled ordering cost is 167,670,000 yen. In the safety stock quantity, the order quantity changes similarly to the demand. Therefore, as shown by the broken line in FIG. 7A, the order quantity varies greatly, and there are cases where a large number of products are delivered at once and the work load is heavy. In addition, since the number of delivered products is large in the safety stock amount, the variation in the work amount at the time of delivery of the ordered product is also large, and it is difficult to shift the workers in the retail store. On the other hand, in the complete leveling of the order quantity, since the number of delivered products is constant, there is no change in the work amount at the time of delivery of the ordered product, and it becomes easy to make a shift of workers in the retail store. However, with the complete leveling of the order quantity, the order cost increases significantly compared to the safety stock quantity. FIG. 7B is a diagram illustrating a change in the inventory amount. In the example of FIG. 7B, the change in the initial stock in each period when the order quantity with the safety stock quantity in FIG. 7A is placed is indicated by a broken line. In the example of FIG. 7B, the change in the initial inventory in each period when the order of the completely equalized order quantity shown in FIG. 7A is performed is indicated by a solid line. In order to complete the order quantity in order to prepare for the later demand, the order quantity is leveled, and therefore, as shown by the solid line in FIG.

  FIG. 8A is a diagram illustrating a change in the order quantity when the order quantity and the order cost based on the conventional safety stock quantity are most emphasized. In the example of FIG. 8A, the change in the order quantity calculated from the safety stock quantity is indicated by a broken line. In the example of FIG. 8A, the change in the order quantity calculated with the highest priority on the order cost is shown by a solid line in accordance with the predicted demand. In the example of FIG. 8A, the ordering cost when the ordering cost is most important is 152,416,860 yen. Thus, when placing importance on ordering cost, the ordering cost is lower than the conventional safety stock quantity. However, the fluctuation of the order quantity becomes large. FIG. 8B is a diagram illustrating a change in the inventory amount. In the example of FIG. 8B, the change in the initial stock in each period when the order quantity with the safety stock quantity in FIG. 8A is placed is indicated by a broken line. Further, in the example of FIG. 8B, the change in the initial inventory in each period when the order quantity with the highest priority on the order cost in FIG. 8A is placed is indicated by a solid line. In this way, when placing importance on ordering costs, the inventory quantity can be increased slightly from the conventional safety inventory quantity.

  The orderer designates the leveling of the order quantity and the importance ratio of the order cost by the track bar 51 of the condition designation screen 50 according to the situation of the retail store. The order quantity determination device 10 outputs a graph indicating the predicted demand and the order quantity calculated according to the designated importance ratio to the display unit 22. For example, a graph of the order quantity with the order cost kept down while leveling the order quantity and leveling the order quantity with the same importance ratio of the order cost is output.

  FIG. 9A is a diagram illustrating an example of an output screen. In the example of FIG. 9A, a graph displayed when leveling the order quantity and the importance ratio of the order cost are the same is shown. For example, the graph of FIG. 9A shows the predicted demand and the order quantity. In this graph, the leveling of the order quantity is specified to be equally emphasized, so the change in the order quantity is less than the change in demand and is a gradual change. In the example of FIG. 9A, the ordering cost is 153,646,900 yen. Thus, the ordering cost is lower than the conventional safety stock quantity. In the example of FIG. 9A, the change in the order quantity is gradual, and the work at the time of product delivery is leveled, so the fluctuation of the work quantity is also reduced, and the shift of workers in the retail store is combined. It becomes easy.

  FIG. 9B is a diagram illustrating a change in the inventory amount. In the example of FIG. 9B, the change in the initial stock in each period when the order quantity with the safety stock quantity in FIG. 7A is placed is indicated by a broken line. Further, in the example of FIG. 9B, the change in the initial stock in each period when the order of the order quantity of FIG. 9A is performed is indicated by a solid line. In this way, the stock quantity when the order quantity is leveled and the importance ratio of the order cost is the same can be increased to several times the conventional safety stock quantity.

  Thus, the order quantity determination device 10 can determine the order quantity according to the leveling of the order quantity and the importance ratio of the order cost.

[Process flow]
Next, the flow of the order quantity determination process in which the order quantity determination device 10 determines the order quantity will be described. FIG. 10 is a flowchart illustrating an example of the order quantity determination process. This order quantity determination process is executed at a predetermined timing, for example, when the order quantity is specified and the importance ratio of the order cost is designated by the track bar 51 of the condition designation screen 50, and the process start is designated. .

  As shown in FIG. 10, the prediction unit 42 uses a linear Gaussian state space model based on the demand performance information 31 to generate a demand generated with a predetermined probability for a product to be ordered for a predetermined period of the ordering target, Predict (S10). The prediction unit 42 stores the prediction result in the demand prediction information 32 (S11).

  The calculation unit 43 sets the weighting coefficient k to the objective function of Expression (39) according to the designation of the track bar 51 on the condition designation screen 50 (S12). The calculation unit 43 solves the optimization problem of the objective function of the equation (39) with the restriction on the order quantity shown in the equation (36) and the limitation on the inventory amount shown in the equation (38) as constraints. In step S13, the order quantity of each product is calculated. The calculation unit 43 stores the calculated order quantity for each period in the order quantity data 33 (S14).

  The output unit 44 outputs a graph indicating the demand stored in the demand prediction information 32 and the order quantity stored in the order quantity data 33 to the display unit 22 (S15), and ends the process.

  As described above, the order quantity determination device 10 according to the present embodiment accepts a demand forecast value of a product. The order quantity determination device 10 calculates the order quantity of the product that minimizes the sum of the index value related to the change in the order quantity of the product and the index value related to the order cost of the product based on the received demand prediction value. Thereby, the order quantity determination apparatus 10 can determine the order quantity according to the actual situation of the ordering work.

  Further, the order quantity determination device 10 calculates the order quantity of the product that minimizes the sum of the weighted values of the index value related to the change in the order quantity of the product and the index value related to the order cost of the product. Thereby, the order quantity determination apparatus 10 can determine the order quantity according to the actual situation of the ordering work by setting the weight according to the actual situation of the ordering work.

  The order quantity determination device 10 calculates the order quantity of the product by solving the optimization problem of the objective function that minimizes the total. Thereby, the order quantity determination apparatus 10 can determine the optimal order quantity.

  Further, the order quantity determination device 10 accepts designation of whether to place importance on order quantity leveling or order cost. The order quantity determination device 10 calculates the order quantity of the product that minimizes the index value related to the change in the order quantity of the product and the index value related to the order cost of the product by weighting according to the received designation. Thereby, the order quantity determination apparatus 10 can determine the order quantity according to the received designation | designated.

  Further, the order quantity determination device 10 outputs a demand forecast value of the product and a graph showing the calculated order quantity. Thereby, the order quantity determination apparatus 10 can show the order quantity corresponding to the demand predicted by the orderer. The orderer can evaluate the validity of the demand and order quantity predicted from the output graph.

  Further, the order quantity determination device 10 predicts the upper limit demand generated with a predetermined probability using a linear Gaussian state space model. The order quantity determination apparatus 10 calculates the order quantity of the product that satisfies the upper limit demand by solving the optimization problem of the objective function, with the constraint that the product will not run out of stock at the upper limit of the predicted demand. To do. As described above, the order quantity determination device 10 can reduce the calculation amount by predicting the upper limit demand generated with a predetermined probability and calculating the order quantity. That is, the order quantity determination device 10 can reduce the calculation quantity because it only calculates the order quantity for only one pattern of the upper limit demand generated with a predetermined probability in the probability distribution of demand. Further, by making the stockout not occur as a restriction condition, it is possible to obtain an order quantity that suppresses occurrence of opportunity loss.

  Although the embodiments related to the disclosed apparatus have been described so far, the disclosed technology may be implemented in various different forms other than the above-described embodiments. Therefore, another embodiment included in the present invention will be described below.

  For example, in the above-described embodiment, a case has been described in which the designation of whether to place importance on order quantity leveling or order cost is accepted, but the present invention is not limited to this. For example, whether to place importance on order quantity leveling or ordering cost may be fixedly set in advance.

  In the above-described embodiment, the case where the calculated order quantity is output as a graph on the screen has been described. However, the present invention is not limited to this. For example, the output unit 44 may output the calculated order quantity to an external device. For example, the output unit 44 may construct the automatic ordering system by outputting the calculated order quantity for the next period to a system that manages the ordering of goods.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific state of distribution / integration of each device is not limited to the one shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. For example, the processing units of the collection unit 40, the reception unit 41, the prediction unit 42, the calculation unit 43, and the output unit 44 may be appropriately integrated. Further, the processing of each processing unit may be appropriately separated into a plurality of processing units. Further, all or any part of each processing function performed in each processing unit can be realized by a CPU and a program analyzed and executed by the CPU, or can be realized as hardware by wired logic. .

[Order quantity determination program]
The various processes described in the above embodiments can also be realized by executing a program prepared in advance on a computer system such as a personal computer or a workstation. Therefore, in the following, an example of a computer system that executes a program having the same function as in the above embodiment will be described. FIG. 11 is a diagram illustrating a computer that executes an order quantity determination program.

  As shown in FIG. 11, the computer 300 includes a CPU (Central Processing Unit) 310, an HDD (Hard Disk Drive) 320, and a RAM (Random Access Memory) 340. These units 300 to 340 are connected via a bus 400.

  The HDD 320 stores in advance an order quantity determination program 320a that performs the same functions as the collection unit 40, reception unit 41, prediction unit 42, calculation unit 43, and output unit 44 described above. Note that the order quantity determination program 320a may be appropriately separated.

  The HDD 320 stores various information. For example, the HDD 320 stores various data used for determining the OS and the order quantity.

  Then, the CPU 310 reads the order quantity determination program 320a from the HDD 320 and executes it, thereby executing the same operation as each processing unit of the embodiment. That is, the order quantity determination program 320a performs the same operations as the collection unit 40, the reception unit 41, the prediction unit 42, the calculation unit 43, and the output unit 44.

  The order quantity determination program 320a is not necessarily stored in the HDD 320 from the beginning.

  For example, the program is stored in a “portable physical medium” such as a flexible disk (FD), a CD-ROM, a DVD disk, a magneto-optical disk, or an IC card inserted into the computer 300. Then, the computer 300 may read and execute the program from these.

  Furthermore, the program is stored in “another computer (or server)” connected to the computer 300 via a public line, the Internet, a LAN, a WAN, or the like. Then, the computer 300 may read and execute the program from these.

DESCRIPTION OF SYMBOLS 10 Order quantity determination apparatus 21 Input part 22 Display part 23 Storage part 24 Control part 30 Product information 31 Demand performance information 32 Demand forecast information 33 Order quantity data 40 Collection part 41 Reception part 42 Prediction part 43 Calculation part 44 Output part

Claims (6)

A storage unit for storing a demand forecast value for a period including a plurality of order times for the product ;
Using the demand forecast value of the period stored in the storage unit as a constraint condition, a first function whose value increases as the order quantity of the product at the plurality of order times varies, and the plurality of order times The order quantity of the product at the plurality of order times is calculated by solving an optimization problem that minimizes an objective function obtained by weighting and adding the second function indicating the order cost of the product at a predetermined ratio. A calculation unit;
An order quantity determination device characterized by comprising: A designation receiving unit for accepting designation of whether order quantity is important or leveling of order quantity;
The calculating unit, according to claim 1, characterized in that to calculate the order quantity of the product to minimize the first function and the second function weighted in accordance with the designation accepted by the designation accepting unit order quantity determination device according to. Order quantity determination device according to claim 1 or 2, characterized by further comprising an output unit for outputting a graph showing the order quantity calculated by the calculation unit and the forecast value of the product. The demand forecast value of the product is an upper limit demand generated with a predetermined probability using a linear Gaussian state space model,
The calculating unit, as a constraint condition that the stock of items in the upper limit of the expected demand does not occur, any claim 1-3, characterized in that to calculate the order quantity of the product to meet the demand of the upper The order quantity determination device according to claim 1. Computer
Using a demand forecast value stored in the storage unit for a period including a plurality of order times for a product as a constraint condition, the value increases as the order quantity of the product at the plurality of order times varies. And solving the optimization problem that minimizes the objective function obtained by weighting and adding the second function indicating the ordering cost of the product at the plurality of order times at a predetermined ratio. A method for determining an order quantity , comprising: executing a process of calculating an order quantity of the product. On the computer,
Using a demand forecast value stored in the storage unit for a period including a plurality of order times for a product as a constraint condition, the value increases as the order quantity of the product at the plurality of order times varies. And solving the optimization problem that minimizes the objective function obtained by weighting and adding the second function indicating the ordering cost of the product at the plurality of order times at a predetermined ratio. order quantity determination program characterized by executing a process for calculating the order quantity of the product.

Images