JP2019079208A - Inventory scale estimation device, energy consumption scale estimation device, cost estimation device, and program - Google Patents

Abstract

To enable a necessary inventory scale to be quantitatively estimated even without achievement data.SOLUTION: A component demand prediction distribution calculation unit 220 calculates a component demand prediction distribution of a final product to be predicted in a prediction period on the basis of a demand prediction value in a prediction period about each of a plurality of final products to be produced, a demand prediction value in a prediction period of the final product to be predicted among the plurality of final products, and a required number of components necessary to produce the final product to be predicted. A by-procurement-period component inventory scale prediction unit 230 predicts a component inventory scale required to produce the final product to be predicted in a prediction period on the basis of a preset safety factor according to an allowable stockout rate of the final product, the calculated component demand prediction distribution, and a predetermined required procurement period necessary to procure the component.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an inventory size estimation device, an energy consumption scale estimation device, a cost estimation device, and a program.

  In the production of parts or products according to the customer's demand, the inventory size is the value obtained by adding the number of safety stocks to the demand forecast value during the procurement time. Conventionally, as a calculation method of the number of safety stocks, for example, as shown in FIG. 13, in Patent Document 1, (1) demand forecast value and demand actual value of the product are acquired from a database, and (2) variance of demand is calculated. ((1) below), (3) cases with uneven distribution of demand are corrected, and (4) the product procurement time is based on the variation of demand against the safety factor based on the allowable shortage rate of the product The value obtained by multiplying the square root value of is calculated as the number of products in stock (the following formula (2)).

  Moreover, also in patent document 2, the method similar to the above is used.

JP, 2005-309770, A JP 2004-359415 A

  The calculation of inventory scale and peak power consumption requires estimation of demand / production variation, whereas the prior art demand / production variation estimation requires actual demand / production data.

  In other words, when factory design, planning, production preparation, and production start-up where actual data are not available or insufficient, it is not possible to estimate stock size etc. quantitatively because it is not possible to estimate variations in demand / production. There was a problem.

  This problem is expected to become even greater in the future, as product diversification is a trend in the world, and with the increase in the number of types of products and parts, the fluctuation of demand is expected to increase. In addition, considering the case where a module of a road where advanced construction and management are desired is to be produced at a factory, factors such as suspension of construction due to rainfall or decline in efficiency are expected, and this is considered as inventory size. It is expected to have an impact. Such a form is not considered in the prior art.

  The present invention has been made to solve the above-mentioned problems, and it is an object of the present invention to provide an inventory size estimation device and program capable of quantitatively estimating the required inventory size even without actual data. .

  Another object of the present invention is to provide an energy consumption scale estimation device capable of quantitatively estimating the required energy consumption scale even without actual data.

  Another object of the present invention is to provide a cost estimation device capable of quantitatively estimating the required cost even without actual data.

  In order to achieve the above object, the inventory size estimation device of the present invention is provided with a forecasting period for each of a plurality of final products to be produced (estimate period of inventory scale, generally, lead time is forecasting period). The demand forecast value of), the demand forecast value of the forecast period of the final product to be forecasted among the plurality of final products, and the required number of components required for the production of the final product to be forecasted And a component part demand forecast distribution calculation unit for calculating a demand forecast distribution of the component part of the final product to be forecasted based on the forecast period, and an allowable shortage of the final product set in advance. Prediction pair in the prediction period based on the safety factor according to the rate, the demand forecast distribution of the component calculated, and the predetermined procurement required period for procuring the component Is configured to include a, a component inventory scale prediction unit to predict the component inventory size needed to produce the final product to be. Here, the final product can be regarded as a component of the 0-dimensional hierarchy and can be a prediction target.

  In the program according to the present invention, the computer further comprises: a demand forecast value of a forecast period for each of a plurality of final products to be produced, and a forecast period of the final product to be forecast among the plurality of final products. Demand forecast distribution of the component of the final product to be forecasted in the forecasting period based on the demand forecast value and the required number of components required to produce the final product to be forecasted The component demand forecast distribution calculation unit for calculating the safety factor, the safety factor set in advance according to the allowable shortage rate of the final product, the demand forecast distribution of the component calculated, and the predetermined. The stock size of the component required to produce the final product to be forecasted is predicted in the forecast period based on the required procurement period for procuring the component. Is a program to function as a component stock-scale prediction unit that.

  According to the inventory scale estimating apparatus and program of the present invention, the component part demand forecast distribution calculation unit calculates the demand forecast value of the forecast period for each of the plurality of final products to be produced, and the plurality of final products among the plurality of final products Based on the demand forecast value of the forecast period of the final product to be forecasted and the required number of component parts required for the production of the final product to be forecasted, the forecasted period will be forecasted The demand forecast distribution of the component parts of the final product is calculated, and the component inventory scale prediction unit sets in advance a safety factor according to the allowable shortage rate of the final product, and the calculated component parts It is necessary to produce the final product to be forecasted in the forecasting period, based on the demand forecast distribution of and the predetermined procurement required period for procuring the component. To predict the stock scale of the serial components.

  Thus, the demand forecast distribution of the component parts of the final product to be forecasted is calculated in the forecast period, the demand forecast distribution of the component parts, and the predetermined time required for procuring the component parts, and Estimate the required inventory size quantitatively even without actual data, by predicting the inventory size of components required to produce the final product to be forecasted in the forecast period based on it can.

  Further, in the inventory scale estimation device of the present invention, the component part demand forecast distribution calculation unit calculates a demand forecast distribution of the component parts for each of the component parts of the final product to be forecasted, and the component parts The stock size prediction unit can predict the stock size of the component for each of the components of the final product to be predicted.

  In addition, the inventory scale estimating device of the present invention adjusts the production of the final product against the influence of the shipping inhibiting factor that inhibits the shipment of the final product to be predicted and the shipping inhibiting factor, which are preset. Distribution that predicts the size of the waiting for shipment of the component based on a production adjustment model which is a model for performing the process and the required number of components required to produce the final product to be predicted The component part inventory scale prediction unit further includes a component part shipment wait distribution distribution calculation unit that calculates a shipment distribution prediction distribution, and the component part inventory scale prediction unit is a prediction target based on the demand distribution of the component parts and the shipment distribution prediction distribution. Inventory sizes of the components of the final product.

  Further, in the inventory scale estimation device of the present invention, the component part demand forecast distribution calculation unit calculates the demand forecast value of the final product to be forecasted with respect to the sum of the demand forecast values for each of the plurality of final products. The demand forecast distribution of the component parts can be calculated based on a binomial distribution model representing whether to select the final product to be forecasted using the ratio of.

  Further, the inventory scale estimation device of the present invention further includes a component demand forecast distribution overlapping unit, and each of the plurality of final products is set as a forecast target, and the component demand forecast distribution calculating unit calculates the plurality of final products For each of the components of the final product, the demand forecast distribution of the components is calculated for each of the components, and the component demand forecast distribution overlapping unit includes the component for each of the components. When there are a plurality of products, the demand forecast distribution of the component calculated for each of the final products is superimposed, and the component stock scale prediction unit is configured to calculate the component calculated for each of the components. The forecasted demand distribution or the forecasted demand distribution of the superimposed component, and the predetermined duration required for procuring the component, in the forecast period. That, it is possible to forecast inventory scale of the components needed to produce a plurality of final product.

  Further, in the inventory scale estimation device of the present invention, the component part demand forecast distribution calculation unit may further consider a supply plan of the final product in the forecast period, which is determined in advance, as a forecast target in the forecast period. The demand forecast distribution of the components of the final product can be calculated.

  An energy consumption scale estimation device according to the present invention produces the component based on the above-described inventory scale estimation device and the demand forecast distribution of the component in a forecast period corresponding to a predetermined energy estimate period. Estimate period-by-period demand forecast distribution calculation unit for calculating the energy demand forecast distribution necessary for the component, the estimate period-by-period demand forecast distribution of the component, and information of equipment required to produce the component; Estimated energy consumption distribution by period to calculate the consumption energy distribution, which is a distribution that represents energy consumption, and the energy consumption scale, based on the information on the energy consumption required for the equipment to produce the component parts And a calculation unit.

  According to the energy consumption scale estimation device of the present invention, the energy demand forecast distribution necessary for producing the component is calculated based on the demand forecast distribution of the component in the forecast period corresponding to the predetermined energy estimate period. Based on the forecasted demand distribution according to the estimated period of the component, the information on the facility required to produce the component, and the information on the energy consumption required in the facility to produce the component By calculating the consumption energy distribution, which is a distribution representing energy consumption, and the energy consumption scale, it is possible to quantitatively estimate the necessary energy consumption scale without actual data.

  The cost estimation apparatus of the present invention calculates inventory costs based on the above energy consumption scale estimation apparatus, the inventory size, and an inventory cost master storing the cost necessary to manage the inventory. A calculation unit is configured to include an energy cost calculation unit that calculates an energy cost based on an energy cost master representing a relationship between the consumption energy distribution and an energy consumption amount and a cost.

  According to the cost estimation device of the present invention, the inventory cost is calculated based on the inventory size and the inventory cost master storing the cost required to manage the inventory, and the energy consumption distribution, the energy consumption amount, and the like. By calculating the energy cost based on the energy cost master representing the relation of costs, it is possible to quantitatively estimate the necessary cost even without actual data.

  As described above, according to the inventory scale estimation device and program of the present invention, the demand forecast distribution of the component parts of the final product to be forecasted is calculated in the forecast period, and the demand forecast distribution of the component parts and Based on the specified procurement requirement period for procuring component parts, it is possible to forecast the stock size of component parts necessary to produce the final product to be forecasted in the forecast period, Even without the data, it is possible to quantitatively estimate the required inventory size.

  Further, according to the energy consumption scale estimation device of the present invention, the energy demand forecast distribution necessary for producing the component based on the demand forecast distribution of the component in the forecast period corresponding to the predetermined energy estimate period Calculating the forecasted demand distribution by component of the period, information of equipment required to produce the component, and information of energy consumption required by the equipment to produce the component By calculating the consumption energy distribution, which is a distribution representing energy consumption, and the energy consumption scale based on the above, it is possible to quantitatively estimate the necessary energy consumption scale without actual data. can get.

  Further, according to the cost estimation device of the present invention, the inventory cost is calculated based on the inventory size and the inventory cost master storing the cost necessary for managing the inventory, and the energy consumption distribution, the energy consumption By calculating the energy cost based on the energy cost master representing the relationship between the amount and the cost, it is possible to quantitatively estimate the necessary cost without the actual data.

It is an image figure showing stock scale calculation concerning an embodiment of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS It is a block diagram showing the structure of the inventory scale estimation apparatus based on the 1st Embodiment of this invention. It is a figure showing an example of composition of a final product. It is a figure showing an example of a demand forecast value. It is a figure which shows an example of a shipping inhibition factor table and a rainfall history. It is a figure showing an example of demand forecast distribution. It is a flowchart which shows the content of the stock scale estimation processing routine in the 1st Embodiment of this invention. It is a block diagram showing the structure of the cost estimation apparatus which concerns on the 2nd Embodiment of this invention. It is a figure showing an example of composition of an end product, and composition of equipment. It is a flow chart which shows the contents of the cost estimate processing routine concerning a 2nd embodiment of the present invention. It is a flow chart which shows the contents of the cost estimate processing routine concerning a 2nd embodiment of the present invention. It is a flow chart which shows the contents of the cost estimate processing routine concerning a 2nd embodiment of the present invention. It is an image figure showing the conventional stock scale calculation.

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

<Overview of the embodiment of the present invention>
Product diversification is a trend in the world, and as the types of products and parts increase, it is expected that demand variation will increase, and it will be important to estimate demand variation.

  There is no actual demand data at the time of planning, preparation for production, and start of production, and it is not possible to estimate the variation in demand from the actual data. This is because the size of the safety stock is calculated from the fluctuation amount of demand calculated from the demand / performance data. Moreover, although there is demand forecast data, it is forecast as an average value, not as a distribution.

  On the other hand, the variation in the demand for each product / part is determined by the variation in whether or not the customer selects the corresponding product, the influence of the sales promotion, etc., and can be estimated from this influence model.

  In the embodiment of the present invention, as shown in FIG. 1, the variation of the demand is estimated not from the actual data but from the demand distribution model. By adopting this configuration, it is possible to estimate the variation of the demand at the stage of planning, factory design, production preparation, production start-up, etc. where there is no or insufficient demand-performance data.

  As a result, stock size including safety stock size, peak power consumption, effect of leveling strategy, etc. can be estimated even at a stage where there is not enough demand data. In addition, it is possible to estimate the stock size including the awaiting shipment considering the shipping impediment factor even at the stage where the actual data is not sufficient.

  The estimation of the size of the safety stock at the stage when there is not sufficient demand and performance data is useful for making decisions on the size of the stock, layout design, etc., and the estimation of peak power consumption is useful for setting the contract power etc. For example, in road unit production at a factory, there are two factors, such as unit size and mass, and fluctuations in shipping inventory size due to the site construction being influenced by the weather, which causes uncertainty in demand and shipping. It is particularly useful because the effect of the prior estimation of inventory size considered is large.

<Configuration of Inventory Scale Estimation Device According to First Embodiment of the Present Invention>
The configuration of the inventory scale estimating apparatus according to the first embodiment of the present invention will be described with reference to FIG. FIG. 2 is a block diagram showing the configuration of the inventory scale estimating apparatus according to the first embodiment of the present invention.

  The inventory scale estimation device 10 calculates the inventory scale of the part type of each component of the portion constituting the final product, which is necessary to produce the final product (for example, a road unit) for a predetermined lead time. Here, in the present embodiment, it is assumed that the final product P whose number of product types is n. Further, as shown in FIG. 3, it is assumed that there are m kinds of component types in the component parts of the part Q of the final product P.

  The inventory scale estimation device 10 is configured by a computer including a CPU, a RAM, and a ROM storing a program for executing an inventory scale estimation process routine described later, and is functionally configured as follows: ing.

  As shown in FIG. 2, the inventory scale estimation device 10 according to the embodiment of the present invention includes an input unit 100, a final product demand forecast value calculation unit 110 by procurement period, a product demand forecast information DB 120, and component parts demand. Predicted value calculation unit 130, product master DB 140, final product demand state acquisition unit 150, demand state model DB 160, component parts demand distribution model acquisition unit 170, demand distribution model DB 180, component parts shipment waiting forecast distribution calculation The unit 190 includes a shipping inhibition factor DB 200, a production adjustment model DB 210, a component demand forecast distribution calculation unit 220, a component inventory scale prediction unit 230 by procurement period, a procurement master DB 240, and an output unit 250. Configured

  The input unit 100 receives, for each of a plurality of final products to be produced, an input of a safety factor according to the allowable shortage rate of the final product and a safety factor according to the space shortage rate.

Specifically, for the final product P, the input unit 100 accepts the probability of shortage and the safety factor (α _i , i = 1 to n) determined from the probability of shortage and the percentage of shortage of space and the percentage of shortage. Accepts an input of a safety factor (γ _i , i = 1 to n) determined by

  The final product demand forecast value calculation unit 110 by procurement period calculates the demand forecast value of the lead period of each of a plurality of final products to be produced. The procurement period is an example of the forecast period. That is, the estimated period of stock size, generally the procurement period, is the forecast period.

  Specifically, first, the final product demand forecast value calculation unit 110 by procurement period acquires the demand forecast information stored in the product demand forecast information DB 120.

  Here, as shown in the following Table (1), the demand forecast information is created in the upstream planning / planning phase, for example, in time units in which each product type of the final product P is produced It includes a production date and a demand forecast value.

  The demand forecast value by the upstream decision-making is mostly forecasted by a value such as an average value (point forecast is made in most cases, and the forecast by the value is also premised in this embodiment.

  Further, in the present embodiment, the time unit of the demand forecasting information is the production date, but the unit of the demand forecasting information may be a unit such as year, month, day, or time.

  Here, when a plurality of time units coexist in the demand forecast information, the final product demand forecast value calculation unit by procurement period 110 uses a forecast of a unit representing the shortest period in the acquired demand forecast information. This is because there is a risk that information such as a trend may be lost if a long period is adopted.

  Then, based on the demand forecast information, the final product demand forecast value calculation unit by procurement period 110 calculates the demand in the procurement period (procurement unit), for example, the demand for 4 hours in the case of procurement every 4 hours, every 8 hours If it is the procurement of, the demand for 8 hours is calculated as the demand forecast value of the final product.

For example, in the example shown in Table (1), it is assumed that daily prediction is made by higher-order decision, and procurement every eight hours (once a day) (that is, the prediction period is one day). The demand forecast value of the product type P0001 is 300 (forecast demand value) × 1 (time interval: day unit) × 1 (for forecast period: day unit) = 300 (demand forecast value _P0001 ).

  The product demand forecast information DB 120 stores a demand forecast value in advance.

  The component part demand forecast value calculation unit 130 calculates, for each of the plurality of final products to be produced, the demand forecast value of the final product calculated by the final product demand forecast value calculation unit 110 according to the procurement period, and the final product Based on the required number of components required for production, the demand forecast value of each of the components of the final product is calculated.

  Specifically, first, the component part demand forecast value calculation unit 130 acquires, from the product master, the product type of the final product as a parent ID and the part type whose part ID is Q and the required number.

  Here, as shown in the following Table (2), the product master is composed of a parent ID which is a product type of the final product, a part ID, a part type, and a required number.

The component demand forecast value calculation unit 130 then calculates the component type of the component in the procurement period from the demand forecast value _i of the product type _i (i = 1,..., N) of the final product in the procurement period and the required number. Demand forecast values _{i, j} of _{i, j} (j = 1,..., m) are calculated (FIG. 4).

For example, the forecasted demand values _{P0001 and Q0001} of the procurement period (one day) of the part category Q0001 used for the product category _P0001 are 300 (demand forecast value _P0001 ) x 1 (required number) = 300 (demand forecast value _{P0001, Q0001} ).

  The product master DB 140 stores a product master in advance.

  The final product demand state acquisition unit 150 acquires the final product demand state based on a predetermined supply plan of the final product in the procurement period.

Specifically, first, the final product demand state acquisition unit 150 determines that the component kind _{i, j} of the component for which production is assumed is the product kind _i of the final product when it is sold as a part of the final product. Get the demand state _i . The time to be sold is, for example, the time obtained by adding the production lead time and the distribution lead time on the day of production.

  The demand state is a state of factors that affect the distribution of demand for product types. In the present embodiment, the case where sales promotion for each product type affects the distribution of demand will be described. Specifically, final product demand state acquisition unit 150 displays a sales promotion plan indicating whether or not the product type stored in the sales promotion plan table of demand state model DB 160 is a sales promotion target, and the demand state table of demand state model DB 160 A demand state table indicating demand states under predetermined conditions stored is acquired.

As shown in the following table (3), the sales promotion planning table is configured to include product types, production months, and values indicating whether or not sales are to be promoted. Here, the state of the sales promotion in August is acquired with August as the time when the part kind _{i, j} is sold as a part of the product kind _i of the final product. The time may be a production day, a production week, or the like.

  Here, if the sales promotion target, the value of the sales promotion is 1, otherwise it is 0. In Table (3), since the sales promotion value is 0 for all n product types of the product P, it is not a sales promotion target.

  As shown in the following table (4), the demand state table includes the demand state and the condition to be the demand state.

The final product demand state obtaining unit 150, from the acquired demand state and demand state table to obtain the demand state _i of product varieties _i. For example, in the example shown in Table (4), the normal state not affected by the sales promotion is acquired as the demand state _i under the condition that the sales promotion value of all product types is 0.

  Here, an example was shown in which the factor determining the demand distribution was promoted as a factor, but the factors of the demand state may be categories of products such as consumable goods and durable goods, general-purpose goods and non-general-purpose goods, seasons and social situations It may be a complex factor including

Then, the final product demand state acquisition unit 150, from the demand state of the product varieties _i of the final product, to obtain the components of the component varieties _i, demand state of _{j i,} a _j. The demand state _{i, j} is the same as the demand state _{i of the} final product. For example, in the above example, since the demand state _i was normal, the demand state _{i, j} of the part type _{i, j} is also acquired as normal.

  The demand state model DB 160 stores in advance a sales promotion plan and a demand state table.

Components demand distribution model acquisition unit 170, based on the demand state final product demand status acquisition unit 150 acquires, parts varieties _{i, j} of demand distribution model _i, obtains the _j.

Specifically, the component part demand distribution model acquisition unit 170 acquires the demand distribution model table in which the demand state and the distribution model are associated from the demand distribution model DB 180, and the final product demand state acquisition unit 150 acquires the demand model The demand distribution model corresponding to the state _{i, j} is acquired. An example of a demand distribution model is shown in the following Table (5).

Here, i is the product type ID, j is the component type ID, n is the total number of product types, req_n _{i, j} is the required number of component types _{j at} a part of product type _i , μ _i is product type _i Average of demand forecast (expected value), μ _{i, j} is average of demand forecast of component type _j as component of product type _i (expected value), σ _i ² is variation of demand forecast of product type _i , σ _{i , J} ² are variations in the demand forecast of the component type _j as a component of the product type _i .

Also, σ _i and σ _{i, j} are standard deviations of the component type _j as components of the product type _i and the product type _i , Demand _i is the demand forecast value of the target procurement period of the product type _i , p _i is the product type The demand ratio of _i (the demand ratio of the part type _j as a component of the product type _i is the same).

  As shown in Table (5), when the demand state is normal, the distribution models of Equation (3) and Equation (4) are acquired.

Here, it supplements about the distribution model of Formula (3) and Formula (4). In the example of Table 5 above, in the normal case, that is, in the case where there is no disturbance such as sales promotion, the distribution of the demand for the final product type is determined by the variation in product type selection of the consumer, and the target product type It is expressed by a binomial distribution model whether or not _i is selected. At this time, the average and the variation are calculated by equation (3). Further, the distribution of the demand of the component types _{i, j} constituting the product type _i is calculated by the equation (4). In the above example, the demand states of normal and sales promotion are shown, but the demand state and the model may be learned from past history data.

  Also, when the demand state is not normal, an arbitrary distribution model is acquired. For example, in the case of sales promotion, the variation of the binomial distribution model in the normal case multiplied by a coefficient less than 1 (for example, 0.8) is acquired as the variation of the distribution model when the demand state is not normal May be Since it is predicted that demand will increase, it is considered that the dispersion of the distribution model in this case is smaller than the dispersion of the distribution model in the normal case. As described above, even when the demand state is adopted other than the sales promotion, an arbitrary distribution model can be acquired according to the demand state.

  The demand distribution model DB 180 stores a demand distribution model table in advance.

  The component shipment waiting prediction distribution calculation unit 190 is a model for adjusting the production of the final product against the influence of the shipping inhibition factor which inhibits the shipment of the final product to be forecasted and the factor of the shipping inhibition which are preset. Based on the production adjustment model, and the required number of components required to produce the final product to be predicted, calculate the shipping waiting forecast distribution, which is a distribution in which the size of the component waiting for shipping is forecasted. .

Specifically, the component shipment waiting prediction distribution calculation unit 190 is based on the shipping inhibition factor that is a factor that inhibits shipment such as rainfall and the production adjustment model for partially adjusting the influence of the shipping inhibition factor. Then, the shipping waiting predicted distribution _{i, j} , which is a distribution for predicting the shipping waiting of the component parts _{i, j to} the delivery destination, is calculated.

  In order to explain the factors affecting shipping, we will show an example of producing a road module at a factory, shipping it and installing it on the site. Considering such a case, termination of construction due to rainfall, freezing, construction machine failure or the like, or a decrease in efficiency, etc. are assumed. That is, it is expected that the shipment of the road module plant will be stopped or the shipment volume will decrease and the inventory for the shipment of the plant will increase, because the rainfall etc. will be a factor to inhibit the shipment.

  In addition, there may be a case where equipment in the downstream process / factory breaks down and the shipment volume decreases. In this case, equipment failure in the downstream plant / process is a factor that inhibits shipment. These shipping impediment factors occur stochastically, and the inventory size awaiting shipping as a result is calculated as a distribution.

For example, as shown in FIG. 5, when the shipping inhibiting factor is rainfall, the distribution is calculated from the data of the rainfall history (the table on the right of FIG. 5). At this time, there are also cases where inventory increases due to shipping inhibition factors are adjusted by production adjustment, and component types of components after adjustment according to a production adjustment model (for example, Table (6) below) that is a model for performing adjustment _The distribution waiting prediction distribution _{i, j} of _{i, j} is calculated.

Here, if the number of awaiting shipments is smaller than the threshold, no production adjustment is made, and all the effects of shipment inhibition appear as an increase in the awaiting shipments. On the other hand, if the number of stocks waiting for shipment is equal to or more than the threshold, the production is adjusted so that the increase in waiting for shipment becomes “the amount of increase for shipment waiting without adjustment × β”. The shipping waiting forecast distribution _{i, j} can be used to estimate the size of stock waiting for shipping.

  In addition, if the final product is a vehicle, a disaster or the like can be considered as a factor inhibiting shipping.

  The shipping inhibition factor DB 200 stores a shipping inhibition factor table in advance.

  The production adjustment model DB 210 stores a production adjustment model in advance.

  The component part demand forecast distribution calculation unit 220 calculates the demand forecast value of the procurement period for each of the plurality of final products to be produced and the demand forecast value of the procurement period of the final product to be forecast among the plurality of final products. The demand forecast distribution of the component parts of the final product to be forecasted in the procurement period is calculated based on and.

Specifically, the component part demand forecast distribution calculation unit 220 acquires the demand forecast values _{i and j} of the component types _{i and j in} the procurement period calculated by the component part demand forecast value calculation unit 130 and the component part demand distribution model acquisition Based on the demand distribution model _{i, j} acquired by the unit 170 _, the demand forecast distribution _{i, j} of the part type _{i, j in} the procurement period is calculated (FIG. 6).

  The component inventory scale prediction unit 230 for each procurement period selects a preset safety component according to the allowable shortage rate of the final product, the predicted demand distribution of the calculated component, and the predetermined component. The stock size of the component parts required to produce the final product to be forecasted in the lead time is predicted based on the lead time required for procurement.

  In addition, the component inventory size prediction unit 230 for each procurement period estimates the required inventory size of the component, the safety factor according to the space shortage allowance ratio of the shipment awaiting component, and the distribution waiting distribution forecast of the calculated component. And a predetermined procurement time required for delivery of a shipping product (for example, it is a delivery for company A but is a procurement for company B, so it is called procurement time). Estimate the stock size of parts to be shipped which will be needed to store parts to be shipped that are subject to forecasting during the procurement period.

  In addition, when there is no shipping impediment factor, the component part inventory size prediction unit 230 by procurement period may be configured to predict the inventory size without considering the shipping waiting forecast distribution.

Specifically, first, the component inventory scale prediction unit 230 by procurement period superimposes the demand forecast distribution of the component types _{i and j} calculated for each product type _i for each component type _j , and is used for all product types The demand forecast distribution _j of each component type to be calculated is calculated. In this example, since the part types Q0001 to Q0003 are used in only one product type respectively, the demand forecast distribution of FIG. 6 becomes the demand forecast distribution of Q0001 to Q0003.

The safety stock size of each part type _{j in} the procurement period is calculated by the following equation (5) with reference to a procurement master table (for example, the following table (7)) indicating the relationship between the part type and the required procurement time. Be done.

Here, safety_inventry _j is the safety stock number of breed _j , α _j is the safety factor of breed _j , procureent LT _j is the procurement time for breed _j , σ _j is the standard deviation (variance) of demand for breed _j Show. Moreover, [x] shows the minimum integer greater than or equal to x.

Further, the stock size including the safety stock of each part type _{j in} the procurement period is calculated by the following equation (6).

Here, mean_inventory _j indicates the average stock amount of the variety j, min_inventory _j indicates the minimum stock amount of the variety j, and max_inventory _j indicates the maximum stock amount of the variety j. Further, Demand _j is a demand forecast value of the target procurement period of the part type _j . However, it is a value in the out-of-stock probability to allow, and may become an inventory amount larger than the maximum stock amount according to the out-of-stock probability.

  Here, although the production uncertainty such as defect occurrence and equipment failure is not taken into consideration, it is also possible to expand the equation to take these into consideration. Further, in the present embodiment, the stock scale is calculated starting from the demand forecast of the final product, but the stock scale may be calculated starting from the demand forecast of the component parts.

  Procurement master DB 240 stores a procurement master table in advance.

  The output unit 250 outputs the stock size predicted by the component stock size prediction unit 230 classified by supply period.

<Operation of Inventory Scale Estimation Device According to First Embodiment of the Present Invention>
Next, with reference to FIG. 7, the stock size estimation processing routine of the stock size estimation apparatus 10 of the present embodiment will be described.

  First, in step S100, for each of a plurality of final products to be produced, the input of a safety factor according to the allowable shortage rate of the final product and the allowable space shortage rate of parts awaiting shipment is received.

  In step S110, the final product demand forecast value calculation unit by procurement period calculation unit 110 acquires the demand forecast information stored in the product demand forecast information DB 120.

  In step S120, the final product demand forecast value calculation unit by procurement period calculation unit 110 calculates the demand forecast value of the final product in the procurement period for each final product based on the demand forecast information.

  In step S130, i = 1. Here, i is a variable representing the id of the final product, and the number of final products is n. Further, j is a variable representing the id of the part type of the component, and the number of part types is m.

In step S140, the component demand forecast value calculation unit 130 acquires, from the product master, the component types _{i, j} of the product type _i of the final product and the required number of component types _{i, j} .

In step S150, the component part demand forecast value calculation unit 130 calculates the demand forecast value _i for each part type _{i, j in} the procurement period from the demand forecast value _i for the product type _i of the final product in the procurement period and the required number _. Calculate _j .

  In step S160, the final product demand state acquisition unit 150 acquires the demand state of the final product based on a predetermined supply plan of the final product in the lead period.

In step S170, the final product demand state acquisition unit 150, from the demand state of the product varieties _i of the final product, to obtain each component varieties _i, demand state of _{j i,} a _j.

In step S180, the component demand distribution model acquisition unit 170 acquires the demand distribution models _{i, j} of the component types _{i, j} based on the demand states _{i, j} acquired in step S170.

In step S190, the component part shipping wait prediction distribution calculation unit 190 calculates the demand forecast values _{i, j} calculated in step S150, the shipping inhibiting factors that are factors that inhibit shipping such as rainfall _, and the effects of the shipping inhibiting factors. Based on the production adjustment model for partial adjustment, the shipping waiting forecast distribution _{i, j} , which is a distribution for forecasting the delivery waiting of each part type _{i, j to} the delivery destination, is calculated. In FIG. 7, the process is performed after step S180. However, the process may be performed anywhere after S150 and before S210.

In step S200, the component part demand forecast distribution calculation unit 220 calculates each part type of the procurement period based on the demand forecast values _{i, j} calculated in step S150 and the demand distribution models _{i, j} acquired in step S180. _{i, j} of demand forecasting distribution _i, to calculate the _j.

  In step S210, it is determined whether i <n.

  If i <n (YES in step S210), 1 is added to i, and the process returns to step S140.

  On the other hand, if i <n is not satisfied (NO in step S210), j = 1 is set in step S220.

In step S230, the component inventory scale prediction unit 230 by procurement period superimposes the demand forecast distribution of the component types _{i and j} calculated for each product type _i for each component type _j , and is used for all product types. The demand forecast distribution _j of the part type is calculated.

In step S240, the component inventory scale prediction unit 230 by procurement period refers to the procurement master table to calculate the safety inventory scale of the part type _{j according} to equation (5).

In step S250, the component inventory scale prediction unit 230 by procurement period superimposes the distribution waiting forecast distribution of the part types _{i and j} calculated for each product type i for each part type j, and is used for all product types. The distribution waiting prediction distribution j of each part type is calculated.

  In this case, the distribution forecasted for distribution is assumed to be stock for delivery (shipping stock), and the demand forecast distribution for production is assumed to be stock for procurement (stocking for arrival), but this is not particularly limited. .

  In step S260, the procurement period-based component parts inventory size prediction unit 230 calculates the safe shipment waiting inventory size of the part type j with reference to the procurement master table.

In step S270, the component inventory size prediction unit 230 for each procurement period predicts the inventory size including the safety inventory of each part type _{j in} the procurement period according to the safety inventory size calculated in step S240 and Equation (6). Further, the stock size for waiting for shipment is predicted from the stock size for safe shipment waiting calculated at step S260.

  In step S280, it is determined whether j <m.

  If j <m (YES in step S280), 1 is added to j, and the process returns to step S230.

  On the other hand, when j <m is not satisfied (NO in step S280), in step S290, the output unit 250 outputs all the stock sizes predicted in step S270.

  As described above, according to the inventory scale estimation device of the first embodiment of the present invention, the demand forecast distribution of the component parts of the final product to be forecasted in the forecast period is calculated, and the demand of the component parts is calculated. Based on the forecasted distribution and the predetermined time required for procuring components, the stock size of components needed to produce the final product to be forecasted is forecasted in the forecasting period. By doing this, it is possible to quantitatively estimate the stock size for the required production, even without actual data. Also, similarly, it is possible to quantitatively estimate the stock size for waiting for required shipping.

<Configuration of Cost Estimating Device According to Second Embodiment of the Present Invention>
Next, with reference to FIG. 8, the configuration of the cost estimation apparatus according to the second embodiment of the present invention will be described. FIG. 8 is a block diagram showing the configuration of the cost estimation device 20 according to the second embodiment of the present invention. In addition, about the part which becomes the structure similar to 1st Embodiment, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  The cost estimation device 20 calculates the stock size of the part type of each component of the part constituting the final product for a predetermined procurement period. Here, in the present embodiment, it is assumed that the final product P whose number of product types is n. Further, as shown in FIG. 9, it is assumed that equipment that can be produced is known for each type of part Q. The estimation of the energy related cost may be any of the equipment unit, the line unit, and the factory unit, but in the present embodiment, an example of calculating the equipment unit is shown.

  The cost estimation device 20 is constituted by a computer provided with a CPU, a RAM, and a ROM storing a program for executing an inventory scale estimation processing routine described later, and is functionally configured as follows: There is.

  As shown in FIG. 8, the cost estimation device 20 according to the embodiment of the present invention includes the input unit 100, the final product demand forecast value calculation unit 110 by procurement period, the product demand forecast information DB 120, and the component parts demand forecast. Value calculation unit 130, product master DB 140, final product demand condition acquisition unit 150, demand condition model DB 160, component parts demand distribution model acquisition unit 170, demand distribution model DB 180, component parts shipment waiting forecast distribution calculation unit 190, shipping inhibition factor DB 200, production adjustment model DB 210, component parts demand forecast distribution calculation unit 220, component parts inventory scale prediction unit 230 by procurement period, procurement master DB 240, inventory cost calculation unit 350, inventory Cost master DB 360, demand forecast distribution calculation unit 370 for estimation period, energy consumption distribution calculation for estimation period And parts 380, and production equipment master DB390, and energy consumption master DB 400, an energy cost calculation unit 410, an energy cost master DB 420, constructed and an output unit 430.

  The inventory cost calculation unit 350 calculates the inventory cost for each procurement period, based on the inventory scale and the inventory cost master in which the cost necessary to manage the inventory is stored.

Specifically, the stock cost calculation unit 350 stores the stock size including the safety stock of the part type _j of each lead time predicted by the lead time component parts stock size prediction unit 230 and the stock cost master DB 360 With reference to the stock cost master, the stock cost of each part type _j is calculated for each procurement period. The inventory cost master stores the cost necessary to manage the inventory, and includes the cost corresponding to the part type, for example, as shown in the following Table (8).

For example, if the stock scale of the part type _Q0001 is 50, the cost is 100 × (50 ÷ 100) = 50 according to the stock cost master in Table (8).

  Inventory cost master DB 360 stores an inventory cost master in advance.

  Estimated period demand forecast distribution calculation unit 370 calculates the demand forecasted distribution of component parts in the energy estimation period, based on the demand forecast distribution of the component parts in each procurement period corresponding to the predetermined energy estimation period.

Specifically, the demand forecast distribution calculation unit 370 for estimation period integrates the demand forecast distribution _j (average and variation) of the part type _j in each procurement period calculated by the component stock scale prediction unit 230 for each procurement period. Thus, the demand forecast distribution (average and variation) of the energy estimation period, for example, the part type _j of one year is calculated.

  The estimated period energy consumption distribution calculation unit 380 is required for the estimated period demand distribution of the component parts, information on the equipment required to produce the component parts, and the equipment for producing the component parts. The energy consumption distribution, which is a distribution representing energy consumption, and the energy consumption scale are calculated based on the information on the energy consumption.

Specifically, first, the energy consumption distribution calculating unit by estimation period 380 stores the demand prediction distribution of the part type _{j in} the energy estimation period calculated by the demand prediction distribution by period calculating unit 370 and the production facility master DB 390 and the time (processing time) for processing by the by component varieties _j one unit of production equipment master and production equipment _k, based on the probability that the component varieties are assigned to production facilities (assignment probability), component varieties _j The processing time distribution _{j, k} in the production facility _k of is calculated.

  Here, in the production facility master, for example, as shown in the following table (9), part types, equipment, processing time, and allocation probability are included.

For example, _assuming that the part type _Q0001 is processed (assigned) via the facility F001, F011 or F012, and F021 and the part type _Q0001 is assigned to F001 and F021, the probability is 100%, and F011 Or the probability assigned to F012 is 50%.

Estimated Period energy consumption distribution calculation unit 380, the time for processing by the production equipment _k parts varieties _j a unit included in the production facility master (the treatment time), based on the assignment probability, parts varieties _j The processing time distribution _{j, k} of the energy estimation period in the production facility _k is calculated.

When the processing time distribution _{j, k} of the energy estimation period in all the facilities is calculated for all the component types, the energy consumption distribution calculating unit by estimation period 380 overlaps the processing time distribution _{j, k} of the energy estimation period to calculate the processing time distribution _k in equipment _k procurement period.

Then, the estimated energy consumption distribution calculation unit 380 calculates the processing time distribution _k of the production facility k and the information on the energy consumption required by the facility to produce the component parts stored in the energy consumption master DB 400. Based on a certain energy consumption master, the consumption energy distribution k, which is a distribution representing the energy consumption of the facility _k , and the energy consumption scale k are calculated. Here, the energy consumption scale is the maximum energy consumption scale that allows a certain energy overuse risk.

  Here, the energy consumption master includes, for example, a facility ID and energy consumption as shown in the following Table (10).

  The production facility master DB 390 stores the production facility master in advance.

  The energy consumption master DB 400 stores the energy consumption master in advance.

  The energy cost calculation unit 410 calculates the energy cost based on the energy consumption distribution and the energy cost master that represents the relationship between the energy consumption and the cost.

Specifically, the energy cost calculation unit 410 calculates the energy cost distribution _k of the facility _{k from} the energy consumption distribution _k , the processing time distribution _k, and the energy cost master stored in the energy cost master DB 420.

Here, the energy cost master includes, for example, the cost per unit time on the production date of the facility _k , as shown in the following Table (11).

  Then, the energy cost calculation unit 410 calculates the energy cost distribution by superimposing the energy cost distribution k (k = 1 to the number of production facilities) of the procurement period.

  The energy cost master DB 420 stores the energy cost master in advance.

  The output unit 430 outputs the inventory cost calculated by the inventory cost calculation unit 350 and the energy cost distribution calculated by the energy cost calculation unit 410.

<Operation of Inventory Size Estimating Device According to Second Embodiment of the Present Invention>
Next, the cost estimation processing routine of the cost estimation apparatus 20 of the present embodiment will be described with reference to FIGS. In addition, about the part which becomes the structure similar to 1st Embodiment, the same code | symbol is attached | subjected and description is abbreviate | omitted.

First, in steps S130 to S210 (FIG. 10), the demand forecast distribution of each component type _{i, j} in the procurement period is calculated.

  In step S315, it is determined whether or not the processes in steps S130 to S210 have been performed for all the procurement periods, and if there is a procurement period in which the processes in steps S130 to S210 have not been performed, the process returns to step S130. The processing of steps S130 to S210 is performed for the period. On the other hand, when the processes of steps S130 to S210 have been performed for all of the procurement periods, the process proceeds to step S320.

  In step S320, j = 1.

In step S330, the component inventory scale prediction unit 230 by procurement period superimposes the demand forecast distribution of the component types _{i and j} calculated for each product type _i for each component type _j , and is used for all product types. The demand forecast distribution _j of each part type in the procurement period is calculated.

In step S340, the component inventory scale prediction unit 230 by procurement period refers to the procurement master table to calculate the safety inventory scale of the part type _j of each procurement period by the equation (5).

In step S350, the component inventory size prediction unit 230 for each procurement period predicts the inventory size including the safety inventory of each part type _{j in} each procurement period, using the safety inventory size calculated in step S340 and Expression (6).

In step S360 (FIG. 11), the stock cost calculation unit 350 refers to the stock size of the part type _{j in} each procurement period predicted in step S270 and the stock cost master stored in the stock cost master DB 360, The inventory cost of the part type _j of the procurement period is calculated.

In step S370, estimated by period forecast distribution calculation section 370, the forecast distribution _j components varieties _j in each lead time calculated in step S230, the calculating a forecast distribution component varieties _j energy estimate period.

  In step S380, k = 1. Here, k is a variable representing the number of production facilities.

In step S390, the energy consumption distribution calculating unit by estimation period 380 acquires the probability (allocation probability) that the part type _j of the production equipment master stored in the production equipment master DB 390 is allocated to the production equipment _k .

In step S400, the energy consumption distribution calculating unit by estimation period 380 acquires the time (processing time) required for processing in the production facility _k of one part of the component type _j of the production facility master stored in the production facility master DB 390. Do.

In step S410, the energy consumption distribution calculating unit by estimation period 380 calculates the demand forecast distribution of the part type _j of the energy estimation period calculated in step S370, the allocation probability acquired in step S400, and the processing time acquired in step S410. Based on the above, the processing time distribution _{j, k} at the production facility _k of the part type _j is calculated.

  In step S420, it is determined whether k <the number of production facilities.

  If k <the number of production facilities (YES in step S420), 1 is added to k, and the process returns to step S390.

  On the other hand, if k is not less than the number of production facilities (NO in step S420), it is determined in step S430 whether j is less than or equal to m.

  If j <m (YES in step S420), 1 is added to j, and the process returns to step S230.

  On the other hand, if j <m is not satisfied (NO in step S420), the process proceeds to step S440 in FIG.

  In step S440, k = 1.

In step S450, estimated by period energy consumption distribution calculating unit 380 calculates a processing time distribution _k in equipment _k processing time distribution _j, procurement period obtained by superimposing _k energy estimate period.

In step S460, the energy consumption distribution calculating unit by estimation period 380 calculates the processing time distribution _k of the production facility _k and the energy consumption required for the facility to produce the component parts stored in the energy consumption master DB 400. based on the energy consumption master is information, calculates the consumption energy distribution k is a distribution representing the consumption of energy of the equipment _k, and energy consumption scale k.

In step S470, the energy cost calculation unit 410 calculates the energy cost distribution _k of the equipment _{k from} the energy consumption distribution _k , the processing time distribution _k, and the energy cost master stored in the energy cost master DB 420.

  In step S480, it is determined whether k <the number of production facilities.

  If k <the number of production facilities (YES in step S480), 1 is added to k, and the process returns to step S450.

  On the other hand, if k is not less than the number of production facilities (NO in step S480), the energy cost calculation unit 410 superimposes the energy cost distribution k (k = 1 to the number of production facilities) of the procurement period in step S490. Calculate

  In step S500, the inventory cost calculated by the inventory cost calculation unit 350 and the energy cost distribution calculated by the energy cost calculation unit 410 are output.

  As described above, according to the cost estimation device of the second embodiment of the present invention, the inventory size and the inventory cost master storing the cost necessary for managing the inventory are used to make the inventory available. By calculating energy costs based on energy cost masters that calculate costs and represent the relationship between energy consumption and energy consumption and costs, it is possible to quantitatively estimate the necessary costs without actual data. it can.

  The present invention is not limited to the above-described embodiment, and various modifications and applications can be made without departing from the scope of the present invention.

  For example, the final product can also be regarded as a component of the 0-dimensional hierarchy. That is, the final product can also be predicted as a component.

  Furthermore, although the present invention has been described as an embodiment in which the program is installed in advance, it is also possible to provide the program by storing the program in a computer readable recording medium.

10 stock size estimation device 20 cost estimation device 100 input unit 110 final product demand forecast value calculation unit by procurement period 120 product demand forecast information DB
130 Component Part Demand Forecast Value Calculation Unit 140 Product Master DB
150 final product demand state acquisition unit 160 demand state model DB
170 Component parts demand distribution model acquisition unit 180 Demand distribution model DB
190 Component parts shipping wait forecast distribution calculation unit 200 shipping inhibition factor DB
210 Production adjustment model DB
220 Component parts demand forecast distribution calculation unit 230 Component parts stock scale prediction unit by procurement period 240 procurement master DB
250 output unit 350 inventory cost calculation unit 360 inventory cost master DB
370 Estimated Period Demand Forecast Distribution Calculation Unit 380 Estimated Period Energy Consumption Distribution Calculation Unit 390 Production Equipment Master DB
400 Energy Consumption Master DB
410 Energy Cost Calculator 420 Energy Cost Master DB
430 output unit

Claims (9)

The demand forecast value for the forecast period for each of a plurality of final products to be produced, the demand forecast value for the forecast period for the final product to be forecast among the plurality of final products, and the forecast target for the forecast A component part demand forecast distribution calculation unit that calculates a demand forecast distribution of the component of the final product to be forecasted in the forecast period, based on the required number of components required for production of the final product; ,
The safety factor according to the allowable shortage rate of the final product set in advance, the demand forecast distribution of the calculated component, and the predetermined time required for procuring the component. A component stock scale prediction unit that predicts the stock scale of the component needed to produce the final product to be forecasted in the forecast period based on
Inventory size estimation device including. The component part demand forecast distribution calculation unit calculates a demand forecast distribution of the component parts for each of the component parts of the final product to be forecasted,
The inventory scale estimating apparatus according to claim 1, wherein the component inventory scale predicting unit predicts an inventory scale of the component for each of the components of the final product to be forecasted. A shipment inhibiting factor which inhibits shipment of the final product to be predicted, and a production adjustment model which is a model for adjusting the production of the final product with respect to the influence of the shipment inhibiting factor, which is set in advance; Component component delivery waiting is performed to calculate a delivery waiting distribution, which is a distribution in which the size of the component component for awaiting shipment is predicted based on the required number of component parts required to produce the final product to be predicted Further includes a predicted distribution calculation unit,
The component parts inventory size prediction unit
The inventory scale estimation device according to claim 1 or 2, wherein the inventory scale of the component of the final product to be forecasted is forecasted based on the demand forecast distribution of the component and the forecasted distribution awaiting shipment.   The component part demand forecast distribution calculation unit uses the ratio of the demand forecast value of the final product to be forecasted to the sum of the demand forecast value for each of the plurality of final products as the forecast target The inventory scale estimating device according to any one of claims 1 to 3, wherein the demand forecast distribution of the component parts is calculated based on a binomial distribution model representing whether or not to select a final product. The component demand forecast distribution unit further includes:
Each of the plurality of final products is to be predicted,
The component part demand forecast distribution calculation unit calculates, for each of the plurality of final products, a demand forecast distribution of the component parts for each of the component parts of the final product,
The component demand forecast distribution superimposing unit superimposes, for each of the components, the demand forecast distribution of the component calculated for each of the final products when there are a plurality of final products including the component.
The component part inventory scale prediction unit is configured to calculate the predicted demand distribution of the calculated component or the predicted demand distribution of the superimposed component for each of the components, and the predetermined component. The inventory scale estimation device according to claim 2, wherein the inventory scale of the component parts required to produce the plurality of final products in the forecast period is forecasted based on a procurement requirement period required for procurement.   The component demand forecast distribution calculation unit further considers the supply plan of the final product in the forecast period, which is determined in advance, and predicts the demand for the component of the final product in the forecast period. The stock size estimation device according to any one of claims 1 to 5, which calculates a distribution. An inventory scale estimating device according to any one of claims 1 to 6,
Estimated demand forecast distribution calculation unit for calculating energy demand forecast distribution necessary to produce the component based on the demand forecast distribution of the component in a forecast period corresponding to a predetermined energy estimation period When,
Demand distribution according to the estimated period according to the estimated period of the component, information of equipment required to produce the component, and information of energy consumption required by the facility to produce the component Energy consumption distribution calculation unit according to an estimation period for calculating the consumption energy distribution, which is a distribution representing energy consumption, and the energy consumption scale, based on
Energy consumption scale estimation device including. The energy consumption scale estimating device according to claim 7;
An inventory cost calculation unit that calculates an inventory cost based on the inventory size and an inventory cost master storing costs necessary to manage the inventory;
An energy cost calculation unit which calculates an energy cost based on an energy cost master representing the relation between the consumption energy distribution and an energy consumption amount and a cost;
Cost estimation equipment including: Computer,
The demand forecast value for the forecast period for each of a plurality of final products to be produced, the demand forecast value for the forecast period for the final product to be forecast among the plurality of final products, and the forecast target for the forecast Component part demand forecast distribution calculation unit which calculates demand forecast distribution of the component of the final product to be forecasted in the forecast period, based on the required number of components required for production of the final product, And a safety factor set in advance according to the allowable shortage rate of the final product, the demand forecast distribution of the calculated component, and a procurement requirement for procuring the component, which is determined in advance. It functions as a component parts stock size prediction unit that predicts the stock size of the component parts necessary to produce the final product to be predicted in the prediction period based on the term and period. Because of the program.