JP2006099691A - Production planning support system, production planning support method, production planning support program, and computer-readable recording medium storing production planning support program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a support system which improves the precision of a production plan and which contributes to planning of production considering numerical side in management. <P>SOLUTION: This production planning support system is a system which supports production planning of products, and includes an insufficient stock and overstock probability calculation part (S101 to S110) which calculates probability of insufficient stock and overstock to present and future production input quantity from variables about production and a profit expected value calculation part (S111 to S114) which calculates an expected value of total profit form the probability of insufficient stock and overstock calculated by the insufficient stock and overstock probability calculation part. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a production plan planning support device, a production plan planning support method, a production plan planning support program, and a computer-readable recording medium that records the production plan planning support program, and in particular, the maximum expected value in terms of management numerical values The present invention relates to a production plan planning support apparatus, a production plan planning support method, a production plan planning support program, and a computer-readable recording medium on which the production plan planning support program is recorded.

  The era of mass production of small varieties has passed, and it has long been said to be an era of small-lot production of various varieties or variety-variable production. In such an era, fluctuations in demand for individual product types are significant. For this reason, a big error must be expected in the demand forecast itself.

  However, the probable production may result in an increase in product inventory due to overproduction, or on the other hand, underproduction may fail to meet demand and cause opportunity loss. For this reason, an important issue for future production companies is how to make a production plan that minimizes the risk of fluctuations in demand.

  Various techniques for optimizing manufacturing instructions based on product demand prediction have been proposed as a method of planning production planning in a product manufacturing factory based on demand fluctuation.

  For example, Patent Document 1 discloses a technique for performing optimization processing by obtaining a plurality of evaluation values based on a demand amount simulation and repeating superiority comparison evaluation between the evaluation values. Patent Document 2 discloses a technique for predicting a safety stock quantity of a product in consideration of a prediction error and calculating a production quantity at a factory.

In view of such a background, in the conventional technology including the technology disclosed in Patent Document 1 and Patent Document 2 described above, the demand quantity and the product inventory quantity are statistically evaluated and the variation is taken into consideration. Various production planning techniques have been proposed.
JP 2003-256635 A JP 2000-293582 A

  However, on the other hand, in an actual factory, not only the variation in the external environment such as the demand quantity, but also a variation occurs with respect to the predicted value with respect to the internal factors such as the production lead time and the production non-defective rate. In the prior art, it cannot be said that these internal factors are sufficiently considered.

  In addition, regarding the variation of the product inventory quantity that has been considered in the conventional technology, the cause of the variation is the variation with respect to the predicted value regarding the production lead time and the production non-defective rate. In the conventional technology, there is a problem in prediction accuracy because the prediction is based on grasping the product inventory quantity as a result rather than the prediction that goes into the cause of these variations. For this reason, there was a problem that the accuracy of the production plan deteriorated.

  Manufacturing lead time includes delay due to equipment trouble, change of production priority for each production type, retention between processes due to failure to synchronize between processes, and increase / decrease in in-process inventory provided as a buffer between processes. Variations occur mainly due to the cause. If the manufacturing lead time varies, the stock will be insufficient, and the delivery date will be delayed with respect to the requested delivery date. If the manufacturing lead time varies, it becomes overstocked.

  Further, the non-defective product yield varies due to the main causes such as variations in accuracy of parts used, variations in process conditions, and irregularities in human work. In the case where the manufacturing non-defective product rate varies, the stock will be insufficient, resulting in a shortage with respect to the requested quantity. If the manufacturing yield rate increases, the stock will be overstocked.

  Loss of sales opportunities and credit guarantees occur when there is a delay in delivery due to a shortage of inventory or missing items. On the other hand, if excess inventory occurs, unsold losses, interest burden on inventory assets, inventory management costs, warehouse costs, and inventory depletion costs occur.

  In this way, when planning production, it is necessary to estimate the costs incurred when inventory shortages and excess inventory occur, and optimize the management values. However, even in this aspect, the conventional technology has not been sufficient. For this reason, there has been a problem that it is not possible to make a production plan that fully considers the management figures.

  The present invention has been made to solve the above-mentioned problems, and one of the objects of the present invention is to provide a production planning support device, a production planning support method, and a production that can improve the accuracy of the production planning. To provide a computer-readable recording medium in which a planning support program and a production planning support program are recorded.

  Another object of the present invention is to provide a production plan planning support apparatus, a production plan planning support method, a production plan planning support program, and a production plan planning support program capable of contributing to the planning of a production plan in consideration of management numerical values. It is to provide a recorded computer-readable recording medium.

  In order to solve the above-described problems, according to one aspect of the present invention, the production plan planning support apparatus supports the production plan planning of products. The production planning support device includes a shortage and excess inventory occurrence probability calculation unit and an expected profit value calculation unit. The insufficient excess inventory occurrence probability calculation unit calculates the occurrence probability of insufficient inventory and excess inventory with respect to the current and future production input quantities from the production-related variables. The expected profit value calculation unit calculates an expected value of the total profit based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage and excess inventory occurrence probability calculation unit.

  According to the present invention, the production planning support device calculates the expected value of the total profit based on the occurrence probability of shortage and excess inventory with respect to the current and future production input quantities calculated from the variables related to production. The For this reason, it is possible to support the planning of production plans by estimating the total profit in the event of shortage or excess inventory. As a result, the production planning accuracy can be improved by the production planning support device. In addition, the production planning support device can contribute to the production planning considering the management numerical value.

  Preferably, the variables are past statistical values, current measured values, and future predicted values.

  Preferably, the production planning support device further includes an actual value detection unit that detects a current actual value. Further, the current actual measurement values are the shipment-finished finished product inventory quantity of a product that has been produced and is waiting for shipment, and the in-process inventory quantity of the product between each process and each process.

  Preferably, the production planning support device stores the actual measurement value detected by the actual measurement value detection unit and the probability density of the actual measurement value based on the actual measurement value stored in the actual measurement value storage unit in the past. And a statistical value calculation unit for calculating the statistical value.

  Further, the current actual measurement value is a production lead time between each process of production and each process, and the past statistical value is a probability density of the lead time. Alternatively, the current actual measurement value is the product non-defective product rate in each process of production, and the past statistical value is the probability density of the product non-defective product rate.

  Preferably, the apparatus further includes a predicted value calculation unit that calculates a future predicted value based on the statistical value calculated by the statistical value calculation unit. Further, the current actual measurement value is the product shipment request quantity, the past statistical value is the probability density of the product shipment request quantity, and the future prediction value is the product shipment request quantity prediction value.

  According to these inventions, the production planning support device detects the current actual measurement value relating to production, the detected actual measurement value is stored, and the probability density of the actual measurement value is determined based on the stored actual measurement value. It is calculated as a statistical value, and a future predicted value is calculated based on the calculated statistical value. For this reason, the expected value of total profit is based on the occurrence probability of shortage and excess inventory for the current and future production inputs calculated from current measured values, past statistical values, and future predicted values. Can be calculated.

  Preferably, the production planning support apparatus further includes a shortage excess inventory loss expected value calculation unit and a sales profit expected value calculation unit. The shortage excess inventory loss expected value calculation unit calculates an expected value of loss based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage excess inventory occurrence probability calculation unit. The sales profit expected value calculation unit calculates the expected value of sales profit based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage and excess inventory occurrence probability calculation unit. The expected profit calculation unit calculates the expected value of profit by subtracting the expected loss calculated by the shortage excess inventory loss calculation from the expected sales profit calculated by the expected sales profit calculation. Calculated as the expected value of total profit.

  According to the present invention, the production planning support device calculates the expected value of loss and the expected value of sales profit based on the probability of occurrence of shortage and excess inventory, and calculates from the calculated expected value of sales profit. The expected value of the total profit is calculated by subtracting the expected value of the generated loss. For this reason, it is possible to calculate the expected value of the total profit based on the occurrence probability of insufficient inventory and excess inventory.

  According to another aspect of the present invention, the production planning support method is a method for supporting production planning of a product by a computer. In addition, the production planning support method includes a step of calculating a probability of occurrence of shortage and excess inventory with respect to current and future production input quantities from production-related variables, and a calculated probability of occurrence of shortage and excess inventory. Calculating an expected value of the total profit based on.

  According to this invention, the accuracy of production planning can be improved by the production planning support method. In addition, the production plan drafting support method can contribute to the production planning in consideration of the management numerical value.

  According to still another aspect of the present invention, the production planning support program is a program that supports product production planning. In addition, the production planning support program calculates the shortage and excess inventory occurrence probabilities for the current and future production inputs from the production variables, and the calculated shortage and excess inventory occurrence probabilities. And calculating the expected value of the total profit based on the computer.

  According to this invention, the accuracy of production planning can be improved by the computer-readable recording medium that records the production planning support program and the production planning support program. In addition, the production plan planning support program and the computer-readable recording medium on which the production plan planning support program is recorded can contribute to the planning of the production plan considering the management numerical value.

  Embodiments of the present invention will be described below with reference to the drawings. In addition, the same code | symbol in the figure shows the same or equivalent member, and the overlapping description is not repeated. In the embodiment of the present invention, as an example of supporting production planning of a product, a case of supporting production planning in a manufacturing factory composed of one assembly process and one inspection process will be described. However, the present invention is not limited to this, and the present invention can be applied to support production planning in a manufacturing factory that includes a plurality of processing steps, a plurality of assembly steps, and a plurality of inspection steps.

<First Embodiment>
FIG. 1 is a functional block diagram showing an outline of functions of a production plan planning support apparatus 1 according to an embodiment of the present invention. The production planning support device 1 determines the occurrence probability of a shortage or excess inventory, which is a cause of a shortage or a delivery delay, from the probability density of the production lead time and the probability density of the shipment required quantity predicted value at the product manufacturing factory. Calculate the expected value of total profit. Then, the production plan planning support apparatus 1 supports the planning of the production plan by presenting the calculated total profit to the user and asking for the judgment by the user.

  The manufacturing factory of this embodiment includes an assembly process 10 and an inspection process 30. In the assembly process 10, the product is assembled from parts. In the inspection step 30, the product is inspected for defective products.

  In addition, the product after assembly becomes the inter-process inventory 20 between the assembly process 10 and the inspection process 30. After the inspection process 30 and before the product is shipped, the finished product after inspection is the finished product inventory 40.

  The production planning support device 1 is composed of a computer such as a PC (Personal Computer) or a workstation. The production planning support device 1 includes an assembly process data calculation unit, an inspection process data calculation unit, an inter-process data calculation unit, a finished product data calculation unit, a shortage excess inventory prediction unit, and an expected profit calculation unit. Including.

  A program for realizing the functions of the assembly process data calculation unit, the inspection process data calculation unit, the inter-process data calculation unit, the finished product data calculation unit, the shortage excess inventory prediction unit, and the expected profit calculation unit The functions of each unit are realized by being executed by a central processing unit (CPU).

  The program may be stored in advance in a nonvolatile memory such as a ROM (Read Only Memory) of the computer, or may be read from an external storage device such as a hard disk drive into a RAM (Random Access Memory) of the computer. May be.

  The program may be provided by a recording medium such as a floppy (registered trademark) disk or a compact disk. The program may be provided via a network such as the Internet. Furthermore, the function realized by the program may be provided via a network.

  In the present embodiment, the functions of the above-described units are realized by software such as a program. However, the present invention is not limited to this, and the function of each unit described above may be realized by a hardware circuit.

  The assembly process data calculation unit collects various production data of the assembly process 10 and calculates necessary data. The inspection process data calculation unit collects various production data of the inspection process 30 and calculates necessary data. The inter-process data calculation unit collects various data of the inter-process inventory 20 and calculates necessary data. The finished product data calculation unit collects various data of the finished product inventory 40 and calculates necessary data.

  The shortage and excess inventory forecasting unit is based on the data calculated by the assembly process data calculation unit, inspection process data calculation unit, inter-process data calculation unit, and finished product data calculation unit. Predict the probability of under and over inventory.

  The expected profit value calculation unit calculates the expected value of the total profit based on the shortage inventory and the excess inventory occurrence probability predicted by the shortage and excess inventory prediction unit.

(Description of Assembly Process Data Calculation Unit in First Embodiment)
The assembly process data calculation unit includes an assembly lead time detection unit 131 and an in-process in-process inventory quantity detection unit 132 as parts for collecting various data of the assembly process 10.

  Further, the assembly process data calculation unit includes an assembly lead time storage unit 121 as a part for storing various data collected in the assembly process 10. Furthermore, the assembly process data calculation unit includes an assembly lead time probability density calculation unit 111 and an assembly shipping possible date / time probability density calculation unit 112 as parts for calculating necessary data of the assembly process 10.

  The assembly lead time detection unit 131 detects the lead time of the product assembly process 10 by measuring the input start time and the payout completion time of each product in the assembly process 10.

  As a means of measuring the input start time and payout completion time of each product, for example, by adding a barcode, two-dimensional code, RF (Radio Frequency) tag, etc. to the product, the individually identified product starts to be input. There is a method of automatically recording the time when the point and the payout completion point are respectively passed. As a result, the assembly lead time detector 131 can automatically detect the lead time of the assembly process 10.

  The lead time data of the assembly process 10 of each product detected by the assembly lead time detection unit 131 is accumulated in the assembly lead time storage unit 121. The accumulated lead time data of the assembly process 10 is used by an assembly lead time probability density calculation unit 111 and an overall lead time probability density calculation unit 411 described later.

  The assembly lead time probability density calculation unit 111 calculates the lead time probability density of the assembly process 10 based on the accumulated data of the assembly process 10 lead time accumulated in the assembly lead time storage unit 121. Table 1 is an example of the probability density data of the lead time of the assembly process 10.

  As shown in Table 1, the assembly lead time probability density calculation unit 111 calculates the probability density of the lead time of the assembly process 10 every 10 minutes. For example, the probability density when the lead time is 130 to 140 minutes can be calculated by the formula (quantity of products whose lead time is 130 to 140 minutes in a predetermined period) / (total product quantity in a predetermined period). it can. Note that the predetermined period is preferably a period sufficiently longer than the lead time.

  Since the lead time takes a continuous value, the probability density function is a continuous function. However, it is not always possible to express the function form by a mathematical expression. Further, even when expression by a mathematical expression is possible, it is not always easy to perform numerical analysis using the mathematical expression. For this reason, in this embodiment, the probability density of the lead time is assumed to be a discrete expression as shown in Table 1.

  The probability density data of the lead time of the assembly process 10 calculated by the assembly lead time probability density calculation unit 111 is transferred to the assembly shipping possible date / time probability density calculation unit 112 described later.

  The probability density data of the calculated lead time of the assembly process 10 is temporarily stored in the database, and then transferred from the database to the assembly shipping possible date / time probability density calculation unit 112. May be.

  The in-process in-process inventory quantity detection unit 132 detects the in-process in-process inventory quantity in the assembly process 10 by measuring the input amount and the withdrawal amount of the product in the assembly process 10. For example, the quantity of products that have been put into the assembly process 10 but not paid out at the time of calculating the total profit is detected as an in-process inventory quantity in the assembly process 10.

  As a means for measuring the input amount and the discharge amount of the product, for example, the input amount and the discharge amount of the individually identified product are automatically recorded by attaching a barcode, a two-dimensional code, an RF tag or the like to the product. There are methods. As a result, the in-process in-process inventory quantity detection unit 132 can automatically detect the in-process in-process inventory quantity of the assembly process 10.

  The in-process in-process inventory quantity data in the assembly process 10 detected by the in-process in-process inventory quantity detection unit 132 is transferred to the assembly shipping possible date / time probability density calculation unit 112 described later. However, the present invention is not limited to this, and the in-process inventory quantity data in the assembly process 10 is once stored in the database and then transferred from the database to the assembly shipping possible date / time probability density calculation unit 112. Good.

(Description of the inter-process data calculation unit in the first embodiment)
The inter-process data calculation unit includes an inter-process lead time detection unit 231 and an inter-process in-process inventory quantity detection unit 232 as parts for collecting various data of the inter-process inventory 20. Further, the inter-process data calculation unit includes an inter-process lead time storage unit 221 as a part for storing various collected data of the inter-process inventory 20. Further, the inter-process data calculation unit includes an inter-process lead time probability density calculation unit 211 and an inter-process shipment possible date / time probability density calculation unit 212 as parts for calculating necessary data of the inter-process inventory 20.

  The inter-process lead time detection unit 231 measures the payout completion time of each product from the assembly process 10 and the input start time to the inspection process 30 in the inter-process inventory 20 between the assembly process 10 and the inspection process 30. Thus, the lead time of the inter-process inventory 20 is detected.

  As means for measuring the input start time and the payout completion time of each product, the same method as that of the assembly lead time detection unit 131 described above can be used.

  The lead time data of the inter-process inventory 20 of each product detected by the inter-process lead time detection unit 231 is accumulated in the inter-process lead time storage unit 221. The accumulated lead time data of the inter-process inventory 20 is used by an inter-process lead time probability density calculation unit 211 and an overall lead time probability density calculation unit 411 described later.

  The inter-process lead time probability density calculation unit 211 is based on the accumulated data from the past of the lead time of the inter-process inventory 20 accumulated in the inter-process lead time storage unit 221, and the probability density of the lead time of the inter-process inventory 20 Is calculated. The inter-process lead time probability density calculation unit 211 calculates the lead time probability density of the inter-process inventory 20 every 10 minutes in the same format as the data shown in Table 1.

  The probability density data of the lead time of the inter-process inventory 20 calculated by the inter-process lead time probability density calculating unit 211 is an assembly shipping possible date / time probability density calculating unit 112 and an inter-process shipping possible date / time probability density calculating unit 212 described later. Forwarded to

  Note that the probability density data of the calculated lead time of the inter-process inventory 20 is temporarily stored in the database, and then the assembly shipping possible date / time probability density calculation unit 112 and the inter-process inventory 20 are calculated. It may be transferred to the shipable date / time probability density calculation unit 212.

  The in-process in-process inventory quantity detection unit 232 measures the stock amount from the assembly process 10 and the input quantity into the inspection process 30 in the inter-process inventory 20 between the assembly process 10 and the inspection process 30, thereby inter-process inventory. 20 inventory quantities are detected. For example, the quantity of products paid out from the assembly process 10 at the time of calculating the total profit but not put into the inspection process 30 is detected as the inventory quantity of the inter-process inventory 20.

  As a means for measuring the input amount and the payout amount of the product, the same method as the in-process in-process inventory quantity detecting unit 132 described above can be used.

  The data of the inventory quantity of the inter-process inventory 20 detected by the inter-process in-process inventory quantity detection unit 232 is transferred to the inter-process shipping possible date / time probability density calculation unit 212 described later. However, the present invention is not limited to this, and the stock quantity data of the inter-process inventory 20 may be temporarily stored in the database and then transferred from the database to the inter-process shipable date / time probability density calculation unit 212. .

(Description of Inspection Process Data Calculation Unit in First Embodiment)
The inspection process data calculation unit includes an inspection lead time detection unit 331 and an in-process inventory quantity detection unit 332 as a part for collecting various data of the inspection process 30. Further, the inspection process data calculation unit includes an inspection lead time storage unit 321 as a part for storing various data collected in the inspection process 30.

  Further, the inspection process data calculation unit includes an inspection lead time probability density calculation unit 311 and an inspection shipping possible date / time probability density calculation unit 312 as parts for calculating necessary data of the inspection process 30.

  The inspection lead time detection unit 331 detects the lead time of the product inspection process 30 by measuring the input start time and the payout completion time of each product in the inspection process 30.

  As means for measuring the input start time and the payout completion time of each product, the same method as that of the assembly lead time detection unit 131 described above can be used.

  The lead time data of the inspection process 30 for each product detected by the inspection lead time detection unit 331 is accumulated in the inspection lead time storage unit 321. The accumulated lead time data of the inspection process 30 is used by an inspection lead time probability density calculation unit 311 and an overall lead time probability density calculation unit 411 described later.

  The inspection lead time probability density calculation unit 311 calculates the probability density of the lead time of the inspection process 30 based on the accumulated data from the past of the lead time of the inspection process 30 stored in the inspection lead time storage unit 321. Then, the inspection lead time probability density calculation unit 311 calculates the probability density of the lead time of the inspection process 30 every 10 minutes in the same format as the data shown as Table 1.

  The probability density data of the lead time of the inspection process 30 calculated by the inspection lead time probability density calculating unit 311 includes an assembly shipping possible date / time probability density calculating unit 112, an inter-process shipping possible date / time probability density calculating unit 212, and an inspection. It is transferred to the shipable date / time probability density calculation unit 312.

  Note that the probability density data of the calculated lead time of the inspection process 30 is temporarily stored in the database, and then the assembly shipping possible date / time probability density calculating unit 112, the inter-process shipping possible date / time are stored. You may make it transfer to the probability density calculation part 212 and the inspection shipping possible date / time probability density calculation part 312.

  The in-process inventory quantity detection unit 332 in the inspection process detects the in-process inventory quantity in the inspection process 30 by measuring the input amount and the withdrawal amount of the product in the inspection process 30.

  As means for measuring the input amount and the discharge amount of the product, the same method as the in-process in-process inventory detecting unit 132 described above can be used.

  The data of the in-process in-process inventory quantity in the inspection process 30 detected by the in-process in-process inventory quantity detection unit 332 is transferred to the later-described inspection shipping possible date / time probability density calculation unit 312. The in-process inventory quantity data in the inspection process 30 is temporarily stored in the database, and then transferred from the database to the inspection shipping possible date / time probability density calculation unit 312. Good.

(Description of the finished product data calculation unit in the first embodiment)
The finished product data calculation unit includes a finished product inventory quantity detection unit 431 as a part for collecting various data of the finished product inventory 40. The finished product inventory quantity detection unit 431 detects the quantity of the finished product inventory 40 stored as an inventory before shipment by measuring the amount dispensed from the inspection process 30 and the shipment quantity from the finished product inventory 40. . For example, the quantity of products paid out from the inspection process 30 at the time of calculating the total profit but not shipped is detected as the inventory quantity of the finished product inventory 40.

  As a means for measuring the amount paid out from the inspection process 30 and the quantity shipped from the finished product inventory 40, for example, the products individually identified by attaching a barcode, a two-dimensional code, an RF tag or the like to the product. There is a method of automatically recording the payout amount and shipment quantity. Thus, the finished product inventory quantity detection unit 431 can automatically detect the inventory quantity of the finished product inventory 40.

  The finished product inventory quantity data stored as inventory before shipment detected by the finished product inventory quantity detection unit 431 is transferred to the production input quantity calculation unit 512 and the shortage excess inventory occurrence probability calculation unit 513.

(Summary of data calculation units between processes and processes in the first embodiment)
The assembly shipping possible date / time probability density calculation unit 112 receives the in-process inventory quantity data of the assembly process 10 transferred from the in-process in-process inventory quantity detection unit 132 and the assembly lead time probability density calculation unit 111. The lead time probability density data of the assembly process 10, the lead time probability density data of the inter-process inventory 20 transferred from the inter-process lead time probability density calculation unit 211, and the inspection lead time probability density calculation unit 311 Based on the probability density data of the lead time of the inspection process 30 that has been performed, the in-process inventory in the assembly process 10 ends the assembly process 10 and is shipped after the work in the inter-process inventory 20 and the inspection process 30 Calculate the probability density of the time required to be possible. Table 2 is an example of the probability density data of the shipping date and time of the in-process inventory in the assembly process 10 and the expected value data of the shipping quantity.

  As shown in Table 2, the assembling / shipping date / time probability density calculating unit 112 calculates the probability density of the shipping availability date / time of the in-process inventory in the assembly process 10 every two hours, and the expected value of the available quantity for each shipping date / time. And calculate. In Table 2, for example, the in-process in-process inventory quantity detection unit 132 shows the expected quantity that can be shipped at each date and time when the in-process in-process inventory quantity of the assembly process 10 is detected as 500 units. ing.

  For example, first, a combination of the lead time of the assembly process 10, the lead time of the inter-process inventory 20 and the lead time of the inspection process 30 that can be shipped from 8:00 to 10:00 on January 15 is obtained. Then, by multiplying the probability density corresponding to each step of the obtained combination and the lead time between the steps and adding all the obtained combinations, the in-process in-process inventory of the assembly step 10 is 8 on January 15th. The probability density that can be shipped from 0:00 to 10:00 is calculated. The expected value of the quantity that can be shipped from 8:00 to 10:00 on January 15 in the in-process inventory of the assembly process 10 is calculated according to the calculated probability density in the process of the assembly process 10 here. Calculated by multiplying the in-process inventory quantity by 500 units.

  Since the shippable date and time takes a continuous value, the probability density function is a continuous function. However, it is not always possible to express the function form by a mathematical expression. Further, even when expression by a mathematical expression is possible, it is not always easy to perform numerical analysis using the mathematical expression. For this reason, in this embodiment, the probability density of the shipping date and time is discretely expressed as shown in Table 2.

  The data on the shipping available date / time probability density of the in-process inventory in the assembly process 10 calculated by the assembly shipping available date / time probability density calculating unit 112 and the expected value data of the available quantity for each available shipping date / time are the entire in-process inventory described later. It is transferred to the shipable date / time probability density calculation unit 412.

  However, the present invention is not limited to this, and the calculated probability density data of the in-process in-process inventory of the assembly process 10 and the expected value data of the quantity available for each shipment available date and time are temporarily stored in the database. After being done, it may be transferred from the database to the entire work-in-progress shipping date / time probability density calculation unit 412.

  The inter-process shipping possible date / time probability density calculation unit 212 receives the in-process inventory quantity data of the inter-process inventory 20 transferred from the inter-process in-process inventory quantity detection unit 232 and the inter-process lead time probability density calculation unit 211. Based on the lead time probability density data of the inter-process inventory 20 and the lead time probability density data of the inspection process 30 transferred from the inspection lead time probability density calculation unit 311, The probability density of the time required to be able to ship through the work in the inventory 20 and the inspection process 30 is calculated.

  Then, the inter-process shipable date / time probability density calculating unit 212 uses the same format as the data shown in Table 2 for the probability density of the date / time when the inter-process inventory 20 can be shipped every two hours, and for each shipable date / time. Expected shipment quantity expected value data is calculated.

  The probability density of the shipment possible date and time and the expected value data of the quantity available for shipment of the inter-process inventory 20 calculated by the inter-process shipmentable date and time probability density calculation unit 212 are transferred to the in-process inventory whole shipmentable date and time probability density calculation unit 412 described later. Is done.

  Note that the present invention is not limited thereto, and the calculated probability density of the available shipping date and time and the expected value data of the available shipping quantity are once stored in the database, and then sent from the database to the entire in-process inventory shipping available date / time probability density calculation unit 412. It may be transferred.

  The inspection shipping possible date / time probability density calculation unit 312 receives the in-process inventory quantity data of the inspection process 30 transferred from the in-process inventory quantity detection unit 332 and the inspection lead time probability density calculation unit 311. Based on the probability density data of the lead time of the inspection process 30, the probability density of the time required until the in-process inventory in the inspection process 30 can be shipped through the work in the inspection process 30 is calculated. To do. Then, the inspection shipping possible date / time probability density calculation unit 312 uses the same format as the data shown in Table 2 and the probability density of the date / time when the in-process in-process inventory of the inspection process 30 every two hours can be shipped, and can be shipped. Create expected value data for the quantity available for each date and time.

  The probability density of the in-process in-process inventory in the inspection process 30 calculated by the inspection / shippable date / time probability density calculation unit 312 and the expected value data of the quantity that can be shipped are calculated as follows. 412.

  Note that the present invention is not limited thereto, and the calculated probability density of the available shipping date and time and the expected value data of the available shipping quantity are once stored in the database, and then sent from the database to the entire in-process inventory shipping available date / time probability density calculation unit 412. It may be transferred.

  In this way, each process and the data calculation unit between the processes, the in-process inventory quantity between the processes or processes, the probability density of the lead time calculated from the past lead times between the processes or processes, and the subsequent process Based on the probability density of the lead time, the probability density of the date and time when each process and the in-process inventory between the processes is completed and can be shipped is calculated.

(Description of the shortage and excess inventory prediction unit in the first embodiment)
The shortage and excess inventory forecasting unit includes a total lead time probability density calculation unit 411, a total in-process inventory shipmentable date / time probability density calculation unit 412, a shipment request quantity storage unit 521, a shipment request quantity prediction value calculation unit 511, and a production input A quantity calculation unit 512 and a deficiency excess inventory occurrence probability calculation unit 513 are included.

  The total lead time probability density calculation unit 411 includes the assembly lead time storage unit 121, the inter-process lead time storage unit 221, and the inspection lead time storage unit 321 transferred from the past of the lead times between the processes and the processes. Based on the accumulated data, the probability density of the total lead time until new production input can be shipped is calculated. Then, the total lead time probability density calculation unit 411 calculates the probability density of the total lead time until a new production input product can be shipped every 10 minutes in the same format as the data shown in Table 1.

  For example, a combination of the lead time of the assembly process 10, the lead time of the inter-process inventory 20, and the lead time of the inspection process 30, in which the lead time falls within a certain range of 10 minutes, is obtained. Then, by multiplying the probability density corresponding to each step of the obtained combination and the lead time between the steps, and adding all the combinations, the probability density that leads the lead time to a value in a range of 10 minutes is calculated. Is done.

  The probability density data of the total lead time until the new production input product that can be shipped calculated by the total lead time probability density calculation unit 411 is transferred to the shortage excess inventory occurrence probability calculation unit 513 described later.

  However, the present invention is not limited to this, and the calculated probability density data of the total lead time may be temporarily stored in the database and then transferred from the database to the shortage excess inventory occurrence probability calculation unit 513.

  The in-process inventory whole-shippable date / time probability density calculation unit 412 includes the probability density of the in-process in-process stock and the expected shipmentable quantity expected value data of the in-process in-process inventory of the assembly process 10 transferred from the assembly / shipment possible date / time probability density calculation unit 112; Probability density and shipping quantity expected value data of the shipping date / time of the inter-process inventory 20 transferred from the inter-process shipping date / time probability density calculation unit 212, and the inspection process 30 transferred from the inspection shipping date / time probability density calculation unit 312. From the probability density of shippable date / time of the in-process in-process inventory and the expected shipmentable quantity data, the probability density of shipable date / time and the expected shipmentable quantity of the entire work-in-process inventory in the factory are calculated. Table 3 is an example of the expected value data of the quantity that can be shipped for every shipment date and time of the entire in-process inventory of the factory.

  As shown in Table 3, the in-process inventory whole shipable date / time probability density calculation unit 412 calculates an expected quantity that can be shipped for every shipable date / time in the in-process inventory in the factory every two hours. For example, the expected value of the quantity that can be shipped with the in-process inventory of the assembly process 10, the inter-process inventory 20, and the inspection process 30 being 8:00 to 10:00 on January 14 is added. Thus, it is possible to calculate the expected quantity that can be shipped from 8:00 to 10:00 on January 14 of the entire in-process inventory.

  Since the shippable date and time takes a continuous value, the probability density function is a continuous function. However, it is not always possible to express the function form by a mathematical expression. Further, even when expression by a mathematical expression is possible, it is not always easy to perform numerical analysis using the mathematical expression. For this reason, in this embodiment, the shipping date and time is expressed in a discrete manner as shown in Table 3.

  The expected value data of the available quantity for every available shipment date / time of the entire in-process inventory in the factory calculated by the entire in-process inventory available date / time probability density calculation unit 412 is transferred to the production input quantity calculation unit 512 described later.

  Note that the present invention is not limited to this, and the expected value data of the quantity that can be shipped for each available shipment date and time of the entire in-process inventory of the factory is once stored in the database, and then from the database to the production input quantity calculation unit 512. It may be transferred.

  The shipment request quantity storage unit 521 stores past shipment request quantity data requested to the factory. The accumulated shipment request quantity data is transferred to the shipment request quantity prediction value calculation unit 511.

  The shipment request quantity predicted value calculation unit 511 calculates the probability density of the future shipment request quantity predicted value based on the accumulated data of the shipment request quantity transferred from the shipment request quantity storage unit 521 from the past. Table 4 is an example of probability density data for each shipping date and time of the shipping request quantity prediction value.

  As shown in Table 4, the shipment request quantity predicted value calculation unit 511 calculates the probability density of the predicted value of the shipment request quantity for each shipment date and time every two hours. The quantity requested for shipment increases and decreases depending on the shipping time zone, the shipping date, the shipping day of the week, the sales trend of the product, and the like. For this reason, the predicted value of the requested shipping quantity is calculated, for example, by adding a predetermined weight to each of a plurality of factors such as the shipping time zone, shipping date, shipping day of the week, and the rate of increase / decrease in the shipping requested quantity. can do.

  The probability density data of the shipping requested quantity predicted value calculated by the shipping requested quantity predicted value calculating unit 511 is transferred to the production input quantity calculating unit 512 and the shortage excess inventory occurrence probability calculating unit 513.

  The production input quantity calculation unit 512 includes the shipment request quantity prediction value transferred from the shipment request quantity prediction value calculation unit 511, and each process of the factory transferred from the work-in-progress inventory whole shipment possibility date / time probability density calculation unit 412 and between each process. The production input quantity is calculated from the expected value of the available quantity for each available delivery date and time of the in-process inventory and the finished product inventory quantity transferred from the finished product inventory quantity detection unit 431.

  For example, the production input quantity calculation unit 512 uses the representative value of the shipment request quantity predicted value transferred from the shipment request quantity predicted value calculation unit 511 to calculate each of the factories transferred from the work-in-progress total shipment possible date / time probability density calculation unit 412. Predetermined quantity obtained by subtracting the expected value of the quantity that can be shipped for each process and the in-process inventory of each process from the delivery date and time, and the finished product inventory quantity transferred from the finished product inventory quantity detection unit 431 Calculated as the production input quantity for the period. Then, the production input quantity calculation unit 512 transfers the calculated production input quantity to the shortage excess inventory occurrence probability calculation unit 513.

  The shortage / overstock occurrence probability calculation unit 513 includes the total lead until the production input quantity transferred from the production input quantity calculation unit 512 and the new production input transferred from the total lead time probability density calculation unit 411 can be shipped. From the time probability density data, the probability density of the shipping date and time of a newly produced input product when the quantity calculated by the production input quantity calculation unit 512 is produced is calculated.

  Then, the shortage and excess inventory occurrence probability calculation unit 513 calculates expected value data of the quantity that can be shipped for each date and time when the newly produced input product can be shipped in the same format as the data shown in Table 3. For example, first, a combination of a total lead time that can be shipped from 8:00 to 10:00 on January 14 and a certain period among predetermined periods is obtained. Then, the probability density corresponding to these combinations is multiplied and added for all the obtained combinations, thereby calculating the probability density that the newly produced input product can be shipped. The expected value of the quantity that can be shipped from 8:00 to 10:00 on January 14 is calculated by multiplying the calculated probability density by the production input quantity for a predetermined period.

  Further, the shortage excess inventory occurrence probability calculation unit 513 is transferred from the expected value data of the available quantity for each date and time when the calculated new production input product can be shipped, and the entire in-process inventory available date and time probability density calculation unit 412. Expected value data of the available quantity for each available shipment date / time of the entire in-process inventory of the factory, probability density data of the expected delivery quantity predicted value transferred from the expected delivery quantity forecast value calculation unit 511, and a finished product inventory quantity detection unit From the finished product inventory quantity data transferred from 431, the excess excess inventory occurrence probability is calculated.

  The calculation of the shortage and excess inventory occurrence probability is performed, for example, by the method described with reference to FIG. FIG. 2 is a graph for explaining a method for calculating the probability of occurrence of shortage and excess inventory performed by the production planning support device 1 according to the first embodiment.

  The upper graph in FIG. 2 shows the probability density data of the shipping request quantity predicted value at the target date and time. Here, out of the probability density data of the shipping request quantity predicted value for each shipping date calculated by the shipping request quantity predicted value calculation unit 511 described above, the probability density of the shipping request quantity predicted value at 15:00 on January 15th. The data is shown. The mode value in the upper graph of FIG. 2 is 300 units.

  The middle graph in FIG. 2 shows a graph obtained by subtracting the finished product inventory quantity to be allocated to the predicted value of the requested shipping quantity at this date and time from the upper graph. Here, the distribution quantity of the finished product stock for the shipment of the finished product stock at this date and time is 40 units. As a result, the mode value of the middle graph in FIG. 2 is 260 units.

  Further, the lower graph of FIG. 2 divides the middle graph by the total quantity of the expected value of the quantity that can be shipped for the date and time of the entire in-process inventory of the factory and the newly produced input product. Here, the total quantity of the expected value of the quantity that can be shipped with respect to the date and time of the entire in-process inventory of the factory and the newly produced input product is 270 units.

  In this figure, the area of the upper part, which has a larger quantity than the quantity 270 units that can be shipped for the current date and time of the entire in-process inventory of the factory and the new production input, is less than the quantity 270 units that can be shipped in the shortage inventory occurrence probability. The area of the lower part is the excess inventory occurrence probability.

  The deficient excess inventory occurrence probability data calculated by the deficient excess inventory occurrence probability calculation unit 513 is transferred to the expected deficient excess inventory loss value calculation unit 514 and the expected saleable quantity expected value calculation unit 515.

(Description of Expected Profit Value Calculation Unit in First Embodiment)
The expected profit value calculation unit includes a shortage excess inventory loss expected value calculation unit 514, a sellable quantity expected value calculation unit 515, a sales profit expected value calculation unit 516, and a total profit expected value calculation unit 517.

  The shortage excess inventory loss expected value calculation unit 514 calculates the expected value of loss cost associated with the occurrence of shortage excess inventory based on the shortage excess inventory occurrence probability data transferred from the shortage excess inventory occurrence probability calculation unit 513.

  An example of a method for calculating the expected value of the shortage and excess inventory occurrence quantity will be described with reference to FIG. FIG. 3 is a graph for explaining a method of calculating an expected shortage of excess inventory generated quantity performed by the production plan planning support apparatus 1 according to the first embodiment.

  FIG. 3 is a graph shown in the lower part of FIG. 2, in which the predicted shipment requested quantity is divided in a range of 10 units, centering on the total in-process inventory of the factory and the available production quantity of 270 units for the relevant date and time. It is a graph.

  The occurrence probability of each divided region is equal to the area of each divided region. In the present embodiment, the median value of the divided areas is N units, and the occurrence probability of the divided areas is expressed as P (N). For example, since the median value of the divided areas of 280 or more and less than 290 is 285, the probability of occurrence of this divided area is expressed as P (285).

  At this time, the expected value of the shortage inventory quantity is Σ {(N−270) × P (N)} for the upper part of the available quantity 270 units for the date and time, that is, for all N 270 <N. It is obtained by calculating.

  The expected value of the loss cost associated with the shortage inventory is calculated by multiplying the expected value of the shortage inventory occurrence quantity thus obtained by the sum of the sales opportunity loss cost per vehicle and the credit compensation cost per vehicle.

  Further, the expected value of the excess inventory occurrence quantity is obtained by applying the same method as the above-described calculation of the expected shortage inventory quantity to the lower part of the graph shown in the lower part of FIG. That is, Σ {(270−N) × P (N)} is calculated for all Ns where N <270.

  By multiplying the expected value of the excess inventory generated in this way by the sum of the inventory interest burden per unit, inventory management cost, warehouse cost, and unsold product risk factor cost, the expected loss cost associated with excess inventory A value is calculated.

  The expected value data of the shortage inventory and the cost associated with the occurrence of excess inventory calculated by the expected shortage excess inventory loss calculation unit 514 is transferred to the total profit expectation value calculation unit 517.

  Based on the shortage inventory and excess inventory occurrence probability data transferred from the shortage and excess inventory occurrence probability calculation unit 513, the expected quantity sale value calculation unit 515 calculates the expected quantity that can be sold at the target date and time.

  An example of a method for calculating the expected quantity that can be sold will be described with reference to FIG. In the figure, the sizable quantity is 270 in the portion above the 270 quantity that can be shipped with respect to the date and time. Therefore, Σ {270 × P (N)} is calculated for all N that satisfy 270 <N.

  In addition, in the portion below the 270 units that can be shipped with respect to the date and time, the number that can be sold is N units. Therefore, Σ {N × P (N)} is calculated for all N that satisfy 270> N. Then, the sum of these calculated values, that is, Σ {min (270, N) × P (N)} is set as the expected quantity that can be sold.

  The sellable volume expected value data calculated by the sellable volume expected value calculation unit 515 is transferred to the sales profit expected value calculation unit 516.

  The sales profit expected value calculation unit 516 calculates the sales profit expected value by multiplying the sales possible quantity expected value transferred from the sellable quantity expected value calculation unit 515 by the sales profit per unit. The sales profit expected value data calculated by the sales profit expected value calculation unit 516 is transferred to the total profit expected value calculation unit 517.

  The total profit expectation value calculation unit 517 calculates the loss cost associated with the shortage of excess inventory and excess inventory generated from the shortage excess inventory loss expectation value calculation unit 514 from the sales profit expectation value transferred from the sales profit expectation value calculation unit 516. The expected value of total profit is calculated by subtracting the expected value. The calculated total profit is presented to the user together with each parameter of the calculation process such as the production input quantity.

  With the production planning support device 1 described above, the probability of occurrence of shortage and excess inventory is calculated from the probability distribution of the production lead time and the probability distribution of the predicted quantity requested for shipment at the product manufacturing plant, and the expectation of the total profit A value can be calculated.

(Description of the flow of processing in the production planning support device 1 in the first embodiment)
Next, a specific process of the production plan planning support apparatus 1 in the first embodiment described above will be described with reference to the flowchart shown in FIG. FIG. 4 is a flowchart showing the flow of the production plan planning support process executed by the production plan planning support apparatus 1 in the first embodiment.

  First, the production planning support device 1 executes the process data calculation process A for the inspection process 30 (step S101). FIG. 5 is a flowchart showing a flow of the process-by-process data calculation process A executed as a subroutine of the production plan planning support process in the first embodiment.

  With reference to FIG. 5, a case where the process data calculation process A is executed for the inspection process 30 will be described here. First, the inspection lead time probability density calculation unit 311 reads the past lead time of the inspection process 30 from the inspection lead time storage unit 321 (step S11).

  The lead time of the inspection process 30 is detected by the inspection lead time detection unit 331 every time the target product passes the inspection process 30. Then, the detected lead time of the inspection process 30 is stored in the inspection lead time storage unit 321 by the inspection lead time detection unit 331.

  Next, the inspection lead time probability density calculation unit 311 calculates the probability density of the lead time of the inspection process 30 based on the lead time of the past inspection process 30 read in step S11 (step S12). Further, the in-process inventory quantity detection unit 332 in the inspection process detects the in-process inventory quantity in the inspection process 30 (step S13).

  Then, the inspection lead time probability density calculation unit 311 calculates the probability density of the shipping date and time after all the processes in-process inventory in the inspection process 30 (step S14). Further, the inspection lead time probability density calculation unit 311 calculates an expected value of the quantity that can be shipped for each shipping date and time after all the in-process inventory in the inspection process 30 (step S15). Thereafter, the process returns to the production planning support process.

  Returning to FIG. 4, the production planning support device 1 executes the data calculation process A for each process for the inter-process inventory 20 and the assembly process 10 as in step S101 (steps S102 and S103).

  Next, the in-process inventory whole shipment possibility date / time probability density calculation unit 412 can send out the in-process inventory of the assembly process 10, the inter-process inventory 20, and the inspection process 30 calculated in the processes from step S101 to step S103. Based on the probability density data and the expected value data of the available quantity for each shipment date and time, the probability density data of the available date and time for the entire in-process inventory of the factory and the expected value data of the available quantity for each shipment date and time are calculated (step) S104).

  Next, the finished product inventory quantity detection unit 431 detects the quantity of the finished product inventory (step S105). Further, the shipment request quantity predicted value calculation unit 511 reads the past shipment request quantity from the shipment request quantity storage unit 521 (step S106). Then, the shipping request quantity predicted value calculation unit 511 calculates the shipping request quantity predicted value based on the accumulated data of the shipping request quantity from the past read in step S105 (step S107).

  Next, the production input quantity calculation unit 512 calculates the probability density data of the shipping available date and time of the whole factory in-process inventory calculated in step S104, the expected value data of the available quantity for each shipping date and time, and the shipping calculated in step 107. The production input quantity is calculated from the requested quantity predicted value data and the finished product inventory quantity data detected in step S105 (step S108).

  Next, the total lead time probability density calculation unit 411 reads the accumulated data of the lead time of the assembly process 10 from the past from the assembly lead time storage unit 121. Similarly, the accumulated lead time data of the inter-process inventory 20 from the past is read from the inter-process lead time storage unit 221, and the accumulated lead time data of the inspection process 30 from the past is read from the inspection lead time storage unit 321. Then, the total lead time probability density calculation unit 411 calculates the total lead time probability density based on each read process and lead time data between the processes (step S109).

  Next, the shortage / overstock occurrence probability calculating unit 513 calculates the probability density data of the total lead time calculated in step S109, the production input quantity data calculated in step S108, and the probability of the shipping possible date / time calculated in step S104. Occurrence of shortage and excess inventory from the density data and the expected value data of the quantity available for each shipment date and time, the predicted shipment quantity calculated in step S107, and the quantity of the finished product inventory calculated in step S105 The probability is calculated (step S110).

  Based on the occurrence probability data of the shortage inventory and excess inventory obtained in this way, the shortage and excess inventory loss expected value calculation unit 514 calculates the expected value of the loss cost associated with the shortage inventory and excess inventory (step S111). . In addition, the sellable quantity expected value calculation unit 515 calculates the expected value of the sellable quantity based on the occurrence probability data of the shortage inventory and excess inventory calculated in Step S110 (Step S112).

Then, based on the expected value data of the available quantity calculated in step S112, the sales profit expected value calculation unit 516 calculates the expected value of sales profit (step S113).
Finally, the total profit expectation value calculation unit 517, based on the expected value data of the loss cost associated with the shortage and excess inventory calculated in step S111 and the expected value data of the sales profit calculated in step S113, An expected value of total profit is calculated (step S114).

(Operations and effects of the production planning support device 1 in the first embodiment)
As described above, the production plan planning support apparatus 1 in the first embodiment is an apparatus that supports the planning of a product production plan. The production planning support device 1 includes a deficiency excess inventory prediction unit and a profit expected value calculation unit.

  The shortage and excess inventory forecasting unit determines the probability of occurrence of shortage and excess inventory for current and future production input quantities from production-related variables such as each process or lead time data between processes and in-process inventory quantity data. calculate. The expected profit value calculation unit calculates the expected value of the total profit based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage and excess inventory prediction unit.

  As a result, the production planning support apparatus 1 calculates the expected value of the total profit based on the occurrence probability of the shortage inventory and excess inventory with respect to the current and future production input quantities calculated from the production-related variables. For this reason, it is possible to support the planning of production plans by estimating the total profit in the event of shortage or excess inventory. As a result, the production planning accuracy can be improved by the production planning support device 1. Further, the production planning support device 1 can contribute to the production planning considering the management numerical value.

  The variables related to production are past statistical values, current measured values, and future predicted values. The current actual measurement values are the production lead times detected between the assembly lead time detection unit 131, the inter-process lead time detection unit 231, and the inspection lead time detection unit 331, the lead time between each process, and the finished product inventory quantity detection. The finished product inventory quantity of the product that has been detected by the unit 431 and is waiting for shipment, the in-process in-process inventory quantity detection unit 132, the in-process in-process inventory quantity detection unit 232, and the in-process in-process inventory quantity This is the in-process inventory quantity of each process detected by the detection unit 332 and between the processes, and the requested shipping quantity stored in the shipping requested quantity storage unit 521.

  The past statistical values are obtained from the assembly lead time probability density calculating unit 111 and the process from the lead times accumulated in the assembly lead time storage unit 121, the inter-process lead time storage unit 221 and the inspection lead time storage unit 321, respectively. The lead time probability density calculating unit 211 and the inspection lead time probability density calculating unit 311 and the lead time probability density between the processes and the shipping request quantity predicted value calculating unit 511 The probability density of the quantity.

  The future predicted value is a predicted value of the requested shipping quantity of the product calculated by the shipping requested quantity predicted value calculation unit 511.

  The production planning support device 1 includes an actual measurement value detection unit that detects a current actual measurement value, an actual measurement value storage unit that stores an actual measurement value detected by the actual measurement value detection unit, and an actual measurement value stored in the actual measurement value storage unit. Select a statistical value calculation unit that calculates the probability density of the actual measurement value as a past statistical value, a predicted value calculation unit that calculates a future predicted value based on the statistical value calculated by the statistical value calculation unit, etc. Including.

  The actual measurement value detection unit includes an assembly lead time detection unit 131, an in-process in-process inventory quantity detection unit 132, an inter-process lead time detection unit 231, an inter-process in-process inventory quantity detection unit 232, and an inspection lead time detection unit. 331, in-process in-process inventory quantity detection unit 332, and finished product inventory quantity detection unit 431.

  The actual measurement value storage unit includes an assembly lead time storage unit 121, an inter-process lead time storage unit 221, an inspection lead time storage unit 321, and a shipping request quantity storage unit 521.

  The statistical value calculation unit includes an assembly lead time probability density calculation unit 111, an assembly shipping date / time probability density calculation unit 112, an inter-process lead time probability density calculation unit 211, an inter-process shipping date / time probability density calculation unit 212, , Inspection lead time probability density calculation unit 311, inspection shipment possible date / time probability density calculation unit 312, total lead time probability density calculation unit 411, whole work-in-progress shipment possible date / time probability density calculation unit 412, shipping request quantity predicted value There is a calculation unit 511.

  As the predicted value calculation unit, there is a shipping request quantity predicted value calculation unit 511. Note that the probability density calculation unit described above may calculate a probability density of a certain variable and calculate a future predicted value of the certain variable from the probability density.

  As a result, the production planning support device 1 detects the current actual measurement value related to production, stores the detected actual measurement value, and based on the stored actual measurement value, the probability density of the actual measurement value is used as a past statistical value. A future predicted value is calculated based on the calculated statistical value. For this reason, the expected value of total profit is based on the occurrence probability of shortage and excess inventory for the current and future production inputs calculated from current measured values, past statistical values, and future predicted values. Can be calculated.

  The production planning support device 1 further includes a shortage excess inventory loss expected value calculation unit 514 and a sales profit expected value calculation unit 516. The shortage and excess inventory loss expected value calculation unit 514 calculates the expected value of loss based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage and excess inventory occurrence probability calculation unit 513. The expected sales profit value calculation unit 516 calculates the expected value of the sales profit based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage excess inventory occurrence probability calculation unit 513.

  Then, the expected profit calculation unit calculates the total profit by subtracting the expected loss value calculated by the shortage excess inventory loss expected value calculation unit 514 from the expected sales profit value calculated by the expected sales profit calculation unit 516. Calculate the expected value.

  Thereby, the production planning support device 1 calculates the expected value of loss and the expected value of sales profit based on the occurrence probability of the shortage and excess inventory, and is calculated from the calculated expected value of sales profit. The expected value of total profit is calculated by subtracting the expected value of loss. For this reason, it is possible to calculate the expected value of the total profit based on the occurrence probability of insufficient inventory and excess inventory.

Second Embodiment
FIG. 6 is a functional block diagram showing an outline of functions of the production plan planning support apparatus 1A according to the second embodiment of the present invention. The production planning support device 1A determines the occurrence probability of a shortage or excess inventory that is a cause of a shortage or a delay in delivery date from the probability density of the non-defective product rate and the probability density of the shipment required quantity predicted value at the product manufacturing factory. To calculate the expected value of total profit. Then, the production plan planning support apparatus 1A supports the planning of the production plan by presenting the calculated total profit to the user and asking for judgment by the user.

  Similar to the production plan planning support apparatus 1 in the first embodiment, the production plan planning support apparatus 1A in the second embodiment is completed with an assembly process data calculation unit, an inspection process data calculation unit, and an inter-process data calculation unit. The product data calculation unit, the deficiency excess inventory prediction unit, and the expected profit calculation unit are included.

(Description of Assembly Process Data Calculation Unit in Second Embodiment)
The assembly process data calculation unit includes an assembly lead time detection unit 131, an in-process in-process inventory quantity detection unit 132, and an assembly non-defective product rate detection unit 133 as parts for collecting various data of the assembly process 10.

  Further, the assembly process data calculation unit includes an assembly lead time storage unit 121 and an assembly non-defective product rate storage unit 122 as parts for storing various data collected in the assembly process 10. Further, the assembly process data calculation unit includes an assembly non-defective product rate probability density calculation unit 113 and an assembling / shipping quantity probability density calculation unit 114 as parts for calculating necessary data of the assembly process 10.

  Since assembly lead time detection unit 131 has been described in the first embodiment, description thereof will not be repeated.

  The lead time data of the assembly process 10 of each product detected by the assembly lead time detection unit 131 is accumulated in the assembly lead time storage unit 121. The accumulated lead time data of the assembly process 10 is used by an assembly shipment possible quantity probability density calculation unit 114 and an overall lead time calculation unit 413 described later.

  The assembly non-defective product rate detection unit 133 detects the product non-defective product rate in the assembly process 10. As a means for detecting a non-defective product rate, for example, there is a method in which a product is individually identified by attaching a barcode, a two-dimensional code, an RF tag, etc., and automatically recorded when a defect is determined. Thereby, the assembly non-defective product rate detection unit 133 can automatically detect the non-defective product rate in the assembly process 10.

  The non-defective product rate data in the assembly process 10 detected by the non-assembled product rate detecting unit 133 is accumulated in the non-assembled product rate storage unit 122. The accumulated non-defective product rate data of the assembly process 10 is transferred to an assembly non-defective product rate probability density calculating unit 113 and an overall non-defective product rate probability density calculating unit 415 to be described later.

  The assembly non-defective product rate probability density calculation unit 113 calculates the probability density of the non-defective product rate in the assembly process 10 based on the accumulated data of the non-defective product rate in the assembly process 10 accumulated in the assembly process 10 stored in the past. Table 5 is an example of non-defective product probability density data in the assembly process 10.

  As shown in Table 5, the non-defective product rate probability density calculation unit 113 calculates the probability density of the non-defective product rate in the assembly process 10 every 0.5%. For example, the probability density when the non-defective product rate is 99.5% or more and less than 100.0% is (number of products whose non-defective product rate is 99.5% or more and less than 100.0%) / (predetermined period) The total product quantity) can be calculated.

  Since the yield rate is a continuous value, the probability density function is a continuous function. However, it is not always possible to express the function form by a mathematical expression. Further, even when expression by a mathematical expression is possible, it is not always easy to perform numerical analysis using the mathematical expression. For this reason, in this embodiment, the probability density of the non-defective product rate is expressed in a discrete manner as shown in Table 5.

  The probability density data of the non-defective product rate of the assembly process 10 calculated by the good assembly rate probability density calculating unit 113 is transferred to the assembling / shipping quantity probability density calculating unit 114 described later. However, the present invention is not limited to this, and the calculated probability density data of the non-defective product ratio in the assembly process 10 is once stored in the database, and then transferred from the database to the assembling / shipping quantity probability density calculating unit 114. May be.

  Since the in-process in-process inventory quantity detection unit 132 has been described in the first embodiment, description thereof will not be repeated. The in-process in-process inventory quantity data in the assembly process 10 detected by the in-process in-process inventory quantity detection unit 132 is transferred to the assembly / shipment possible quantity probability density calculation unit 114. However, the present invention is not limited to this, and the in-process inventory quantity data in the assembly process 10 is once stored in the database and then transferred from the database to the assembling / shipping quantity probability density calculation unit 114. Good.

(Description of the inter-process data calculation unit in the second embodiment)
The inter-process data calculation unit includes an inter-process lead time detection unit 231 and an inter-process in-process inventory quantity detection unit 232 as parts for collecting various data of the inter-process inventory 20. Further, the inter-process data calculation unit includes an inter-process lead time storage unit 221 as a part for storing various collected data of the inter-process inventory 20. Further, the inter-process data calculation unit includes an inter-process shipmentable quantity probability density calculation unit 213 as a part for calculating necessary data of the inter-process inventory 20.

  Since the inter-process lead time detection unit 231 has been described in the first embodiment, description thereof will not be repeated. The lead time data of the inter-process inventory 20 of each product detected by the inter-process lead time detection unit 231 is accumulated in the inter-process lead time storage unit 221. The accumulated lead time data of the inter-process inventory 20 is used by an assembly shipment possible quantity probability density calculation unit 114, an inter-process shipment possible quantity probability density calculation unit 213, and an overall lead time calculation unit 413, which will be described later.

  Since the in-process in-process inventory quantity detection unit 232 has been described in the first embodiment, description thereof will not be repeated. The in-process inventory quantity data of the in-process inventory 20 detected by the inter-process in-process inventory quantity detection unit 232 is transferred to the inter-process shipmentable quantity probability density calculation unit 213. However, the present invention is not limited to this, and the in-process inventory quantity data of the inter-process inventory 20 is once accumulated in the database, and then transferred from the database to the inter-process shipmentable quantity probability density calculation unit 213. Good.

(Description of Inspection Process Data Calculation Unit in Second Embodiment)
The inspection process data calculation unit includes an inspection lead time detection unit 331, an in-process in-process inventory quantity detection unit 332, and an inspection non-defective product rate detection unit 333 as parts for collecting various data of the inspection process 30.

  The inspection process data calculation unit includes an inspection lead time storage unit 321 and an inspection acceptable product rate storage unit 322 as parts for storing various data collected in the inspection process 30. Further, the inspection process data calculation unit includes an inspection good product rate probability density calculation unit 313 and an inspection shipment possible quantity probability density calculation unit 314 as parts for calculating necessary data of the inspection process 30.

  Since inspection lead time detection unit 331 has been described in the first embodiment, description thereof will not be repeated. The lead time data of the inspection process 30 for each product detected by the inspection lead time detection unit 331 is accumulated in the inspection lead time storage unit 321. The accumulated lead time data of the inspection process 30 includes an assembly shipment possible quantity probability density calculation unit 114, an inter-process shipment possible quantity probability density calculation unit 213, and an inspection shipment possible quantity probability density calculation unit 314, which will be described later. Used by the lead time calculation unit 413.

  The non-defective product rate detection unit 333 detects the non-defective product rate in the inspection process 30. As a means for detecting the non-defective product rate of the product, the same method as that of the non-defective product rate detecting unit 133 in the assembling process 10 can be used.

  The non-defective product rate data in the inspection process 30 detected by the test non-defective product rate detection unit 333 is accumulated in the test non-defective product rate storage unit 322. The accumulated non-defective product rate data of the inspection process 30 is transferred to an inspection non-defective product rate probability density calculating unit 313 and a comprehensive non-defective product rate probability density calculating unit 415 described later.

  The inspection non-defective product rate probability density calculation unit 313 is based on the accumulated data of the non-defective product rate from the past in the inspection process 30 stored in the inspection good product rate storage unit 322 in the same format as the data shown in Table 5 The probability density of the non-defective product rate in the inspection process 30 every 5% is calculated.

  The probability density data of the non-defective product rate of the inspection process 30 calculated by the inspection non-defective product rate probability density calculating unit 313 includes an assembly shipping possible quantity probability density calculating unit 114, an inter-process shipping possible quantity probability density calculating unit 213, and an inspection. It is transferred to the shipable quantity probability density calculation unit 314.

  The probability density data of the non-defective product ratio of the inspection process 30 calculated is not limited to this, and is stored in the database, and then can be shipped from the database to the assembling / shipping quantity probability density calculating unit 114 and the inter-process shipping. It may be transferred to the quantity probability density calculation unit 213 and the inspection / shippable quantity probability density calculation unit 314.

  Since the in-process inventory quantity detection unit 332 in the inspection process has been described in the first embodiment, description thereof will not be repeated. The in-process in-process inventory quantity data in the inspection process 30 detected by the in-process in-process inventory quantity detection unit 332 is transferred to the later-described inspection shipping possible quantity probability density calculation unit 314. However, the present invention is not limited to this, and the in-process inventory quantity data in the inspection process 30 is once stored in the database, and then transferred from the database to the inspection shipping possible quantity probability density calculation unit 314. Good.

(Description of Completed Product Data Calculation Unit in Second Embodiment)
The finished product data calculation unit includes a finished product inventory quantity detection unit 431 as a part for collecting various data of the finished product inventory 40. Since the finished product inventory quantity detection unit 431 has been described in the first embodiment, description thereof will not be repeated. The finished product inventory quantity data stored as inventory before shipment detected by the finished product inventory quantity detection unit 431 is transferred to the production input quantity calculation unit 512A and the shortage excess inventory occurrence probability calculation unit 513A.

(Summary of data calculation units between processes and processes in the second embodiment)
The assembling / shipping quantity probability density calculation unit 114 includes the in-process in-process inventory quantity data of the assembly process 10 transferred from the in-process in-process inventory quantity detection unit 132 and the assembly process stored in the assembly lead time storage unit 121. 10 lead time data, lead time data of the inter-process inventory 20 transferred from the inter-process lead time storage unit 221, lead time data of the inspection process 30 transferred from the inspection lead time storage unit 321, and assembly Based on the probability density data of the non-defective product rate of the assembly process 10 transferred from the non-defective product rate probability density calculating unit 113 and the probability density data of the non-defective product rate probability density data of the inspection process 30 transferred from the non-defective product rate probability density calculating unit 313. In addition, the entire in-process inventory of the assembly process 10 ends the assembly process 10, and the inter-process inventory 20 and the work in the inspection process 30 are performed. A deliverable date allows shipping Te to calculate the yield rate probability density. Table 6 is an example of the quantity that can be shipped for each non-defective product ratio in the in-process inventory in the assembly process 10 and the probability density data of the non-defective product rate.

  As shown in Table 6, the assembling / shipping quantity probability density calculating unit 114 calculates the quantity that can be shipped with respect to the non-defective product rate every 0.5% at the shipping possible date and time of the entire in-process inventory in the assembly process 10 and the overall non-defective product rate. Probability density data is calculated.

  For example, the possible shipment date and time of the entire in-process inventory in the assembly process 10 can be calculated by adding the average values of the lead times of the assembly process 10, the inter-process inventory 20, and the inspection process 30, respectively. Further, a combination of the non-defective product ratios of the assembling process 10 and the inspecting process 10 in which the overall non-defective product ratio obtained by multiplying the non-defective product ratios of the assembly process 10 and the inspecting process 10 is 99.5% or more and less than 100% is obtained. Then, the probability density of the whole non-defective product rate is calculated by multiplying the probability density of the non-defective product rate of each step of the obtained combination.

  Furthermore, when the overall non-defective product rate is 100%, the entire in-process in-stock quantity in the assembly process 10 is the quantity that can be shipped. Therefore, the quantity that can be shipped when the overall non-defective rate is 99.5% is 199 units by multiplying the quantity that can be shipped when the overall non-defective rate is 100% by 99.5%. Can be calculated. Then, the calculated shipment possible quantity of 199 is set as a shipmentable quantity when the overall non-defective product ratio is 99.5% or more and less than 100%.

  Since the yield rate is a continuous value, the probability density function is a continuous function. However, it is not always possible to express the function form by a mathematical expression. Further, even when expression by a mathematical expression is possible, it is not always easy to perform numerical analysis using the mathematical expression. For this reason, in this embodiment, the probability density of the yield rate is discretely expressed as shown in Table 6.

  The shipment possible quantity data and the probability density data of the non-defective product rate of the entire in-process in-process inventory of the assembly process 10 calculated by the assembly / shipable quantity probability density calculation unit 114 can be shipped as a whole in-process inventory, which will be described later. It is transferred to the quantity probability density calculation unit 414.

  Note that the present invention is not limited to this, and the data on the quantity that can be shipped at the shipping date and time of the entire in-process inventory of the assembly process 10 and the probability density data of the non-defective product rate are once accumulated in the database. It may be transferred from the database to the whole work-in-progress shipmentable quantity probability density calculation unit 414.

  The inter-process shipable quantity probability density calculation unit 213 receives the inter-process in-process inventory quantity data of the inter-process inventory 20 transferred from the inter-process in-process inventory quantity detection unit 232 and the inter-process lead time storage unit 221. Accumulated lead time data of the inter-process inventory 20, accumulated lead time data of the inspection process 30 transferred from the inspection lead time storage unit 321, and inspection process 30 transferred from the inspection good product rate probability density calculation unit 313. Based on the probability density data of the non-defective product rate, the shipping date and time and the non-defective product rate probability density at which the entire inter-process inventory 20 can be shipped through the inter-process inventory 20 and the work in the inspection process 30 are calculated. .

  Then, the inter-process shipmentable quantity probability density calculation unit 213 uses the same format as the data shown in Table 6 to provide the quantity data that can be shipped with respect to the non-defective product rate for every 0.5% of the inter-process inventory 20 at the date of shipment. And probability density data of the yield rate.

  The inter-process shipmentable quantity probability density calculation unit 213 calculates the total shipmentable quantity data of the inter-process inventory 20 and the probability density data of the non-defective product rate to the in-process inventory total shipmentable quantity probability density calculation unit 414 described later. Transferred.

  However, the present invention is not limited to this, and the calculated shipping quantity data of the entire inter-process inventory 20 and the probability density data of the non-defective product rate can be temporarily stored in the database and then shipped from the database as a whole in-process inventory. It may be transferred to the quantity probability density calculation unit 414.

  The inspection shipping available quantity probability density calculation unit 314 includes the in-process in-process inventory quantity data of the inspection process 30 transferred from the in-process in-process inventory quantity detection unit 332 and the inspection process transferred from the inspection lead time storage unit 321. Based on the accumulated data of the lead time of 30 and the probability density data of the non-defective product rate in the inspection process 30 transferred from the inspection non-defective product rate probability density calculation unit 313, the entire in-process inventory in the inspection process 30 is inspected. The shipment possible date and time and the non-defective product rate probability density that can be shipped after completion of the above are calculated.

  Then, the inspection shipmentable quantity probability density calculation unit 314 has the same format as the data shown in Table 6, and the quantity that can be shipped with respect to the non-defective product rate for every 0.5% in the shipmentable date and time of the entire in-process inventory in the inspection process 30. And the probability density data of the yield rate are calculated.

  The shipment possible quantity data of the entire in-process in-process inventory of the inspection process 30 and the probability density data of the non-defective product rate calculated by the inspection shipment possible quantity probability density calculation unit 314 are calculated as follows. Transferred to the unit 414.

  However, the present invention is not limited to this, and the calculated shipmentable quantity data of the entire in-process inventory of the inspection process 30 and the probability density data of the non-defective product rate are once stored in the database, and then stored in the in-process inventory. It may be transferred to the entire inventory shipmentable quantity probability density calculation unit 414.

  In this way, each process and the data calculation unit between the processes, the in-process inventory quantity between the processes or processes, the probability density of the non-defective ratio calculated from the past non-defective ratio of the process, the past between the processes or processes Based on the lead time accumulation data, the past lead time accumulation data of the post-process, and the probability density of the non-defective product ratio calculated from the past non-defective product ratio of the post-process, the entire in-process inventory between each process and the process is completed Then, the probability density of the quantity that can be shipped at the date and time when the product can be shipped is calculated.

(Description of the shortage and excess inventory prediction unit in the second embodiment)
The shortage and excess inventory prediction unit includes an overall lead time calculation unit 413, an in-process inventory total shipmentable quantity probability density calculation unit 414, an overall non-defective product rate probability density calculation unit 415, a shipment request quantity storage unit 521, and a shipment request quantity prediction. A value calculation unit 511, a production input quantity calculation unit 512A, and a shortage excess inventory occurrence probability calculation unit 513A are included.

  The total lead time calculation unit 413 stores accumulated data of past lead times between processes and processes transferred from the assembly lead time storage unit 121, the inter-process lead time storage unit 221, and the inspection lead time storage unit 321. Based on the above, the total lead time data until new production input products can be shipped is calculated.

  For example, the total lead time calculation unit 413 stores the past accumulated data of each process and the lead time between processes stored in the assembly lead time storage unit 121, the inter-process lead time storage unit 221 and the inspection lead time storage unit 321, respectively. Are averaged for each process and between processes, and the total lead time is calculated by adding for each process and between processes.

  The total lead time data until the new production input product calculated by the total lead time calculation unit 413 can be shipped is transferred to the shortage excess inventory occurrence probability calculation unit 513A described later.

  However, the present invention is not limited to this, and the calculated total lead time data may be temporarily stored in the database and then transferred from the database to the insufficient excess inventory occurrence probability calculating unit 513A.

  The comprehensive non-defective rate probability density calculation unit 415 includes the accumulated data from the past of the non-defective product rate of the assembly process 10 transferred from the assembled non-defective product rate storage unit 122 and the inspection process 30 transferred from the inspection non-defective product rate detection unit 333. Based on the accumulated data of the non-defective rate from the past, the probability density data of the total non-defective rate of the whole factory is calculated in the same format as the data shown as Table 5.

  For example, the non-defective product rate in the assembly process 10 and the non-defective product rate in the inspection process 30 are multiplied to obtain a combination of the non-defective product rate in the assembly process 10 and the non-defective product rate in the inspection process 30 within a certain range of the non-defective product rate. Then, the probability density corresponding to the non-defective product rate of each step of the obtained combination is multiplied and added to all the combinations to calculate the probability density in which the overall non-defective product rate falls within a certain range.

  The overall quality rate probability density data of the entire factory calculated by the overall quality rate probability density calculation unit 415 is transferred to the shortage excess inventory occurrence probability calculation unit 513A described later.

  However, the present invention is not limited to this, and the calculated probability density data of the total non-defective rate may be temporarily stored in the database and then transferred from the database to the insufficient excess inventory occurrence probability calculating unit 513A.

  The whole in-process inventory shipmentable quantity probability density calculation unit 414 transfers the data of the quantity available for shipment and the probability of the non-defective product rate at the shipment date / time of the entire in-process in-process inventory in the assembly process 10 transferred from the assembly shipment possibility quantity probability density calculation unit 114. Density data, shipmentable quantity data and non-defective rate probability density data at the shipmentable date / time of the entire interprocess inventory 20 transferred from the interprocess shipmentable quantity probability density calculation unit 213, and inspection shipmentable quantity probability density calculation unit The probability density of the quantity that can be shipped at a certain date and time in the whole in-process inventory at the factory is calculated from the data of the quantity available for shipment and the probability density data of the non-defective product ratio of the entire in-process inventory in the inspection process 30 transferred from 314. calculate. Table 7 is an example of the probability density data of the quantity that can be shipped for each shipping date and time of the entire in-process inventory of the factory, and the expected value of the quantity that can be shipped for each shipping date and time.

  As shown in Table 7, the whole work-in-progress shipmentable quantity probability density calculation unit 414 calculates the probability density with respect to the quantity available for shipment for each shipable date and time of the whole work-in-process inventory in the factory. For example, by adding the probability density for each available quantity of products that can be shipped from 8:00 to 10:00 on January 15 in the entire work-in-process inventory, for each process and between processes, January 15 The probability density is calculated for each quantity that can be shipped from 8:00 to 10:00. Then, the expected value of the shipment quantity of 8:00 to 10:00 on January 15 is obtained by multiplying the shipment quantity and the probability density corresponding to the shipment quantity and adding all the shipment quantities. Is calculated.

  The probability density data of the available quantity for each shipment date and the expected value of the available quantity for each shipment date and time of the whole in-process inventory in the factory calculated by the whole work-in-progress shipmentable quantity probability density calculation unit 414 is the production input quantity described later. The data is transferred to calculation unit 512A.

  However, the present invention is not limited to this, and the probability density data of the quantity that can be shipped for each shipping date and time and the expected value of the quantity that can be shipped for each shipping date and time of the calculated factory in-process inventory are once stored in the database. The data may be transferred from the database to the production input quantity calculation unit 512A.

  Since the shipping request quantity storage unit 521 and the shipping request quantity predicted value calculation unit 511 have been described in the first embodiment, description thereof will not be repeated. The probability density data of the shipping requested quantity predicted value calculated by the shipping requested quantity predicted value calculating unit 511 is transferred to the production input quantity calculating unit 512A and the shortage excess inventory occurrence probability calculating unit 513A.

  The production input quantity calculation unit 512A includes the shipment request quantity prediction value transferred from the shipment request quantity prediction value calculation unit 511, and the factory processes transferred from the work-in-progress whole-stock shipment possible quantity probability density calculation unit 414 and between the processes. The production input quantity is calculated from the available quantity for each shipment date and time of the entire in-process inventory and the finished product inventory quantity transferred from the finished product inventory quantity detection unit 431.

  For example, the production input quantity calculation unit 512A uses the representative value of the shipment request quantity predicted value transferred from the shipment request quantity predicted value calculation unit 511 to calculate each of the factories transferred from the work-in-progress inventory whole shipmentable quantity probability density calculation unit 414. The quantity obtained by subtracting the expected value of the quantity that can be shipped for each shipment date and time of the entire process and the in-process inventory between each process and the finished product inventory quantity transferred from the finished product inventory quantity detection unit 431, Calculated as production input quantity. Then, the production input quantity calculation unit 512A transfers the calculated production input quantity to the insufficient excess inventory occurrence probability calculation unit 513A.

  The shortage excess inventory occurrence probability calculation unit 513A includes the production input quantity transferred from the production input quantity calculation unit 512A and the total lead time data until the new production input transferred from the total lead time calculation unit 413 can be shipped. And the probability density data of the overall non-defective product rate transferred from the general non-defective product probability density calculation unit 415, the new input product when the quantity calculated by the production input quantity calculation unit 512A is input. Shippable date / time and probability density of quantity available for shipment are calculated.

  Then, the shortage and excess inventory occurrence probability calculation unit 513A calculates probability density data of the quantity that can be shipped for each date and time when the newly produced input product can be shipped in the same format as the data shown in Table 7.

  Further, the shortage / overstock occurrence probability calculation unit 513A is transferred from the probability density data of the quantity that can be shipped for each date and time when the calculated new production input can be shipped, and the quantity probability density calculation unit 414 that can be shipped as a whole in-process inventory. The probability density data of the quantity that can be shipped for each shipping date and time of the entire in-process inventory of the factory, the probability density data of the shipping request quantity predicted value transferred from the shipping request quantity predicted value calculation unit 511, and the finished product inventory quantity detection unit 431 From the finished product inventory quantity data transferred from, the excess excess inventory occurrence probability density is calculated.

  The calculation of the shortage / overstock occurrence probability is performed by the method described with reference to FIGS. 7 and 8, for example. FIG. 7 is a first graph for explaining a method of calculating the excess excess inventory occurrence probability performed by the production planning support device 1A according to the second embodiment. FIG. 8 is a second graph for explaining a method of calculating the excess excess inventory occurrence probability performed by the production planning support device 1A according to the second embodiment.

  The upper graph in FIG. 7 shows the probability density data of the shipping request quantity predicted value at the target date and time. Here, out of the probability density data of the shipping request quantity predicted value for each shipping date calculated by the shipping request quantity predicted value calculation unit 511 described above, the probability density of the shipping request quantity predicted value at 15:00 on January 15th. The data is shown. The mode value in the upper graph of FIG. 7 is 300 units.

  The middle graph of FIG. 7 shows a graph obtained by subtracting the finished product inventory quantity to be allocated to the predicted value of the requested shipping quantity at this date and time from the upper graph. Here, the distribution quantity of the finished product stock for the shipment of the finished product stock at this date and time is 40 units. As a result, the mode value of the middle graph in FIG. 7 is 260 units.

  In the lower graph of FIG. 7, the middle graph is divided in a range of 10 units. The upper graph in FIG. 8 is the same as the lower graph in FIG. With reference to the upper graph of FIG. 8, the graph of the predicted value of the requested shipment quantity is divided in a range of 10 units.

The median of each divided divided regions and N _i stand. Here, the probability that the requested shipment quantity is (N _i −5) or more and less than (N _i +5) is expressed as P ₁ (N _i ). For example, the median of the divided region of less than 290 or more units 280 units, because it is 285 units, the probability of the divided region is expressed as P ₁ (285). At this time, the probability P ₁ (N _i ) is equal to the area in the range of (N _i -5) or more and less than (N _i +5) in the graph of the predicted value of the shipment request quantity.

  The lower graph of FIG. 8 is a graph showing the total value of the shipmentable quantity probability density of the entire in-process inventory of the factory and the shipmentable quantity probability density of the newly produced input product. This graph is divided in the same divided area as the upper graph of FIG.

The median value of each divided region is M _j units. Note that i and j are set so that N _i = M _j when i = j. Here, the probability that the requested shipment quantity is (M _j −5) or more and less than (M _j +5) is expressed as P ₂ (M _j ). At this time, the probability P ₂ (M _j ) is equal to the area in the range of (M _j −5) or more and less than (M _j +5) in the lower graph of FIG.

The occurrence probability of insufficient inventory is obtained by calculating ΣΣ {P ₁ (N _i ) × P ₂ (M _j )} for all i, j where i> j. The occurrence probability of excess inventory is obtained by calculating ΣΣ {P ₁ (N _i ) × P ₂ (M _j )} for all i, j where i <j.

  The shortage / overstock occurrence probability density calculated by the shortage / overstock occurrence probability calculation unit 513A is transferred to the shortage / overstock loss expected value calculation unit 514A and the sellable quantity expected value calculation unit 515A.

(Explanation of Expected Profit Value Calculation Unit in Second Embodiment)
The expected profit value calculation unit includes a shortage excess inventory loss expected value calculation unit 514A, a sellable volume expected value calculation unit 515A, a sales profit expected value calculation unit 516A, and a total profit expected value calculation unit 517A.

  The expected shortage excess inventory loss calculation unit 514A calculates the expected value of the loss cost associated with the occurrence of shortage excess inventory based on the shortage excess inventory occurrence probability density transferred from the shortage excess inventory occurrence probability calculation unit 513A.

The expected value of the shortage of inventory is obtained by calculating ΣΣ {(N _i −M _j ) × P ₁ (N _i ) × P ₂ (M _j )} for all i and j where i> j. It is done.

  The expected value of the loss cost associated with the shortage inventory is calculated by multiplying the expected value of the shortage inventory occurrence quantity thus obtained by the sum of the sales opportunity loss cost per vehicle and the credit compensation cost per vehicle.

Further, the expected value of the excess inventory generation quantity is to calculate ΣΣ {(M _j −N _i ) × P ₁ (N _i ) × P ₂ (M _j )} for all i, j where i <j. Is obtained.

  By multiplying the expected value of the excess inventory generated in this way by the sum of the inventory interest burden per unit, inventory management cost, warehouse cost, and unsold product risk factor cost, the expected loss cost associated with excess inventory A value is calculated.

  The shortage excess inventory loss expectation value calculation unit 514A obtained by the shortage inventory and the loss cost expectation data associated with the excess inventory occurrence is transferred to the total profit expectation value calculation unit 517A.

  The expected saleable quantity calculation unit 515A calculates the expected saleable quantity at the target date and time based on the shortage inventory and excess inventory occurrence probability data transferred from the shortage excess inventory occurrence probability calculation unit 513A.

An example of a method for calculating the expected quantity that can be sold will be described with reference to FIG. Since N _i ≦ M _j , the sellable quantity is N _i , ΣΣ {N _i × P ₁ (N _i ) × P ₂ (M _j )} is set for all i and j where i ≦ j. calculate.

On the other hand, since the sellable quantity when N _i > M _j is M _j , ΣΣ {M _j × P ₁ (N _i ) × P ₂ (M _j )} for all i and j where i> j. Calculate Then, ΣΣ {min (N _i , M _j ) × P ₁ (N _i ) × P ₂ (M _j )}, which is the sum of these calculated values, becomes the expected quantity that can be sold.

  The sellable volume expected value data calculated by the sellable volume expected value calculation unit 515A is transferred to the sales profit expected value calculation unit 516A.

  The sales profit expectation value calculation unit 516A calculates the sales profit expectation value by multiplying the sales profit expectation value transferred from the sellable quantity expectation value calculation unit 515A by the sales profit per unit. The sales profit expected value data calculated by the sales profit expected value calculation unit 516A is transferred to the total profit expected value calculation unit 517A.

  The total profit expectation value calculation unit 517A calculates the loss cost associated with the shortage inventory and the excess inventory occurrence from the shortage excess inventory loss expectation value calculation unit 514A from the sales profit expectation value transferred from the sales profit expectation value calculation unit 516A. The expected value of total profit is calculated by subtracting the expected value. The calculated total profit is presented to the user together with each parameter of the calculation process such as the production input quantity.

  With the above production planning support device 1A, the probability of occurrence of shortage and excess inventory is calculated from the probability distribution of the production lead time and the probability distribution of the predicted quantity requested for shipment in the manufacturing factory of the product, and the expectation of the total profit A value can be calculated.

(Description of the flow of processing in the production planning support device 1A in the second embodiment)
Next, a specific process of the production plan planning support apparatus 1A in the second embodiment described above will be described with reference to the flowchart shown in FIG. FIG. 9 is a flowchart showing a flow of production plan planning support processing executed by the production plan planning support apparatus 1A in the second embodiment.

  First, the production planning support device 1A executes the process data calculation process A for the inspection process 30 (step S201). FIG. 10 is a flowchart showing the flow of the process-by-process data calculation process B executed as a subroutine of the production plan planning support process in the second embodiment.

  Referring to FIG. 10, here, a case where the process data calculation process B is executed for the inspection process 30 will be described. First, the inspection non-defective product rate probability density calculation unit 313 reads the past non-defective product rate of the inspection process 30 from the inspection good product rate storage unit 322 (step S21).

  The non-defective product rate in the inspection process 30 is detected by the test non-defective product rate detection unit 333 every time the target product passes the inspection process 30. The detected non-defective product rate in the inspection process 30 is stored in the inspected non-defective product rate storage unit 322 by the test non-defective product rate detecting unit 333.

  Next, the inspection non-defective rate probability density calculation unit 313 calculates the probability density of the non-defective product rate in the inspection step 30 based on the past non-defective rate in the inspection step 30 read in step S21 (step S22). Further, the in-process inventory quantity detection unit 332 in the inspection process detects the in-process inventory quantity in the inspection process 30 (step S23).

  Next, the inspection shipment possible quantity probability density calculation unit 314 reads the past lead time of the inspection process 30 from the inspection lead time storage unit 321 (step S24). The lead time of the inspection process 30 is detected by the inspection lead time detection unit 331 every time the target product passes the inspection process 30. Then, the detected lead time of the inspection process 30 is stored in the inspection lead time storage unit 321 by the inspection lead time detection unit 331.

  Then, the inspection / shippable quantity probability density calculation unit 314 calculates the shippable date / time and the non-defective product rate probability density after all the processes in-process inventory in the inspection process 30 (step S25). In addition, the inspection shipping available quantity probability density calculation unit 314 calculates the quantity available for each non-defective product rate after the whole process in-process inventory in the inspection process 30 and the probability density of the good product rate (step S26). . Thereafter, the process returns to the production planning support process.

  Returning to FIG. 9, the production planning support device 1A executes the inter-process data calculation process a for the inter-process inventory 20 (step S202). FIG. 11 is a flowchart showing the flow of the inter-process data calculation process a executed as a subroutine of the production plan planning support process in the second embodiment.

  With reference to FIG. 11, the case where the inter-process data calculation process a is executed for the inter-process inventory 20 will be described here. First, the in-process in-process inventory quantity detection unit 232 detects the inventory quantity of the in-process inventory 20 (step S31).

  Then, the inter-process shipmentable quantity probability density calculation unit 213 reads the past lead time of the inter-process inventory 20 from the inter-process lead time storage unit 221 (step S32). The lead time of the inter-process inventory 20 is detected by the inter-process lead time detection unit 231 every time the target product passes through the inter-process inventory 20. The detected lead time of the inter-process inventory 20 is stored in the inter-process lead time storage unit 221 by the inter-process lead time detection unit 231.

  Further, the inter-process shipmentable quantity probability density calculation unit 213 calculates the data of the quantity that can be shipped for each non-defective product rate after all the processes of the inter-process inventory 20 and the probability density of the non-defective product ratio (step S33). Thereafter, the process returns to the production planning support process.

  Returning to FIG. 9, the production planning support device 1A executes the process data calculation process B for the inspection process 30 in the same manner as in step S201 (step S103).

  Next, the total in-process inventory available quantity probability density calculation unit 414 calculates the total in-process in-process inventory of each of the assembly process 10, the inter-process inventory 20, and the inspection process 30 calculated in the processes from step S201 to step S203. Calculate the probability density data of the available quantity for each shipment date and time and the expected value of the available quantity for each shipment date and time from the data on the quantity available for shipment and the probability density data for the non-defective product rate. (Step S204).

  Next, the finished product inventory quantity detection unit 431 detects the quantity of the finished product inventory (step S205). Further, the shipment request quantity predicted value calculation unit 511 reads the past shipment request quantity from the shipment request quantity storage unit 521 (step S206). Then, the shipping request quantity predicted value calculation unit 511 calculates the shipping request quantity predicted value based on the accumulated data of the shipping request quantity from the past read in step S206 (step S207).

  Next, the production input quantity calculation unit 512A calculates the probability density data of the quantity that can be shipped for each shipping date and time of the entire in-process inventory of the factory calculated in step S204, the expected value data for the quantity that can be shipped for each shipping date and time, and step S207. A production input quantity is calculated from the calculated shipment request quantity predicted value data and the finished product inventory quantity data detected in step S205 (step S208).

  Next, the comprehensive non-defective rate probability density calculation unit 415 is based on the accumulated data of the non-defective product rate of the assembly process 10 and the inspection process 30 from the past stored in the assembled non-defective product rate storage unit 122 and the inspected non-defective product rate storage unit 322, respectively. The probability density of the total non-defective product rate is calculated (step S209).

  Also, the total lead time calculation unit 413 includes an assembly process 10, an inter-process inventory 20, and an inspection stored in the assembly lead time storage unit 121, the inter-process lead time storage unit 221, and the inspection lead time storage unit 321, respectively. The total lead time is calculated based on the accumulated lead time data in step 30 (step S210).

  Next, the shortage / overstock occurrence probability calculating unit 513A includes the probability density data of the total non-defective product rate calculated in step S209, the total lead time data calculated in step S210, and the production input quantity data calculated in step S208. , The probability density data of the available quantity for each shipment date and time of the entire factory in-process inventory calculated in step S204, the expected value data of the available quantity for each shipment date and time, and the predicted requested shipment quantity value calculated in step S207 Then, the occurrence probability of the shortage inventory and the excess inventory is calculated from the quantity of the finished product inventory calculated in step S205 (step S211).

  Based on the occurrence probability data of the shortage and excess inventory calculated in step S211, the shortage and excess inventory loss expected value calculation unit 514A calculates the expected value of the loss cost associated with the shortage and excess inventory (step S212). In addition, the expected saleable quantity calculation unit 515A calculates the expected saleable quantity based on the shortage and excess inventory occurrence probability data calculated in step S211 (step S213).

  Then, the expected sales profit value calculation unit 516A calculates the expected value of the sales profit based on the expected value data of the available quantity calculated in step S213 (step S214).

  Finally, the total profit expectation value calculation unit 517A, based on the expected value data of the loss cost associated with the shortage and excess inventory calculated in step S212 and the expected value data of the sales profit calculated in step S214, An expected value of total profit is calculated (step S215).

(Operations and effects of the production planning support device 1A in the second embodiment)
As described above, the production plan planning support device 1A in the second embodiment is a device that supports the planning of a product production plan. Further, the production planning support device 1A includes a deficiency excess inventory prediction unit and a profit expected value calculation unit.

  The shortage and excess inventory forecasting section is based on production variables such as yield rate data for each process, lead time data for each process or process, and in-process inventory quantity data. Calculate the occurrence probability of inventory and excess inventory. The expected profit value calculation unit calculates the expected value of the total profit based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage and excess inventory prediction unit.

  As a result, the production planning support device 1A calculates the expected value of the total profit based on the occurrence probability of the shortage and excess inventory with respect to the current and future production input quantities calculated from the variables related to production. For this reason, it is possible to support the planning of production plans by estimating the total profit in the event of shortage or excess inventory. As a result, the production planning accuracy can be improved by the production planning support device 1A. In addition, the production plan planning support apparatus 1A can contribute to the planning of a production plan in consideration of management numerical values.

  The variables related to production are past statistical values, current measured values, and future predicted values. The current actual measured values are the non-defective product ratios in each production process detected by the assembly good product rate detection unit 133 and the inspection good product rate detection unit 333, the assembly lead time detection unit 131, the inter-process lead time detection unit 231 and the inspection lead. The production process detected by the time detection unit 332 and the lead time between the processes, the finished product inventory quantity of the product waiting for shipment after the production detected by the finished product inventory detection unit 431 is completed, In-process in-process inventory quantity detection unit 132, in-process in-process in-process inventory quantity detection unit 232, and in-process in-process in-process inventory quantity detection unit 332, and in-process inventory quantity of products between each process and shipping request This is the requested shipping quantity stored in the quantity storage unit 521.

  Further, the past statistical values are obtained from the non-defective product rate storage unit 122 and the non-defective product rate storage unit 322, and the non-defective product rate probability density calculating unit 113 and the inspecting non-defective product rate probability density calculating unit 313, respectively. The probability density of the non-defective product rate of each process calculated by the above, and the probability density of the requested shipping quantity calculated by the shipping requested quantity predicted value calculation unit 511.

  The future predicted value is a predicted value of the requested shipping quantity of the product calculated by the shipping requested quantity predicted value calculation unit 511.

  The production planning support device 1A includes an actual value detection unit that detects a current actual value, an actual value storage unit that stores an actual value detected by the actual value detection unit, and an actual value stored in the actual value storage unit. Select a statistical value calculation unit that calculates the probability density of the actual measurement value as a past statistical value, a predicted value calculation unit that calculates a future predicted value based on the statistical value calculated by the statistical value calculation unit, etc. Including.

  The actual measurement value detection unit includes an assembly non-defective product rate detection unit 133, an assembly lead time detection unit 131, an in-process in-process inventory quantity detection unit 132, an inter-process lead time detection unit 231 and an in-process in-process inventory quantity detection unit. 232, an inspection good product rate detection unit 333, an inspection lead time detection unit 331, an in-process in-process inventory quantity detection unit 332, and a finished product inventory quantity detection unit 431.

  The actual measurement value storage unit includes an assembly good product rate storage unit 122, an assembly lead time storage unit 121, an inter-process lead time storage unit 221, an inspection good product rate storage unit 322, and an inspection lead time storage unit 321.

  As the statistical value calculation unit, an assembly non-defective product rate probability density calculation unit 113, an assembling / shipping quantity probability density calculation unit 114, an inter-process shipmentable quantity probability density calculation unit 213, an inspection non-defective product rate probability density calculation unit 313, Inspection shipment possible quantity probability density calculation unit 314, total good product rate probability density calculation unit 415, total lead time calculation unit 413, whole work-in-stock shipment possible quantity probability density calculation unit 414, shipment requested quantity predicted value calculation unit 511 There is.

  As the predicted value calculation unit, there is a shipping request quantity predicted value calculation unit 511. Note that the probability density calculation unit described above may calculate a probability density of a certain variable and calculate a future predicted value of the certain variable from the probability density.

  As a result, the current actual measurement value relating to production is detected by the production planning support device 1A, the detected actual measurement value is stored, and the probability density of the actual measurement value is determined as a past statistical value based on the stored actual measurement value. A future predicted value is calculated based on the calculated statistical value. For this reason, the expected value of total profit is based on the occurrence probability of shortage and excess inventory for the current and future production inputs calculated from current measured values, past statistical values, and future predicted values. Can be calculated.

  The production planning support device 1A further includes a shortage excess inventory loss expected value calculation unit 514A and a sales profit expected value calculation unit 516A. The shortage excess inventory loss expected value calculation unit 514A and the sales profit expectation value calculation unit 516A are the same as the shortage excess inventory loss expectation value calculation unit 514 and the sales profit expectation value calculation unit 516 in the first embodiment, respectively. Has an effect.

<Third Embodiment>
FIG. 12 is a functional block diagram illustrating an outline of functions of the production plan planning support apparatus 1B according to the third embodiment of the present invention. The production planning support device 1B has a shortage of inventory and a shortage of delivery due to a product lead time factor, a product lead time probability density, a product non-defective product rate probability density, and a shipping demand quantity predicted value probability density. And the occurrence probability of excess inventory is calculated, and the expected value of total profit is calculated. Then, the production plan planning support apparatus 1B supports the planning of the production plan by presenting the calculated total profit to the user and asking for the judgment by the user.

  The production plan planning support apparatus 1B in the third embodiment, like the production plan planning support apparatus 1 in the first embodiment, is completed with an assembly process data calculation unit, an inspection process data calculation unit, and an inter-process data calculation unit. The product data calculation unit, the deficiency excess inventory prediction unit, and the expected profit calculation unit are included.

(Description of the assembly process data calculation unit in the third embodiment)
The assembly process data calculation unit includes an assembly lead time detection unit 131, an in-process in-process inventory quantity detection unit 132, and an assembly non-defective product rate detection unit 133 as parts for collecting various data of the assembly process 10.

  Further, the assembly process data calculation unit includes an assembly lead time storage unit 121 and an assembly non-defective product rate storage unit 122 as parts for storing various data collected in the assembly process 10. Further, the assembly process data calculation unit calculates the necessary data of the assembly process 10 as an assembly lead time probability density calculation unit 111, an assembly non-defective product rate probability density calculation unit 113, and an assembly shipment date / time quantity probability density calculation unit. 115.

  Since assembly lead time detection unit 131 has been described in the first embodiment, description thereof will not be repeated.

  The lead time data of the assembly process 10 of each product detected by the assembly lead time detection unit 131 is accumulated in the assembly lead time storage unit 121. The accumulated lead time data of the assembly process 10 is used by an assembly lead time probability density calculation unit 111 and an overall lead time calculation unit 413 which will be described later.

  Since assembly lead time probability density calculation unit 111 has been described in the first embodiment, description thereof will not be repeated. The lead time probability density data of the assembly process 10 calculated by the assembly lead time probability density calculation unit 111 is transferred to an assembly shipment date / time quantity probability density calculation unit 115 described later.

  However, the present invention is not limited thereto, and the calculated lead time probability density data of the assembly process 10 is once stored in the database and then transferred from the database to the assembly shipment date / time quantity probability density calculation unit 115. May be.

  Since the assembled good product rate detection unit 133 has been described in the second embodiment, description thereof will not be repeated. The non-defective product rate data in the assembly process 10 detected by the non-assembled product rate detecting unit 133 is accumulated in the non-assembled product rate storage unit 122. The accumulated non-defective product rate data of the assembly process 10 is transferred to an assembly non-defective product rate probability density calculating unit 113 and an overall non-defective product rate probability density calculating unit 415 to be described later.

  The assembly good product rate probability density calculation unit 113 has been described in the second embodiment, and therefore description thereof will not be repeated. The probability density data of the non-defective product rate in the assembly process 10 calculated by the good assembly rate probability density calculation unit 113 is transferred to the assembly shipment date / time quantity probability density calculation unit 115 described later. However, the present invention is not limited to this, and the calculated probability density data of the non-defective product ratio in the assembly process 10 is once stored in the database and then transferred from the database to the assembly / shipping date / time quantity probability density calculation unit 115. May be.

  Since the in-process in-process inventory quantity detection unit 132 has been described in the first embodiment, description thereof will not be repeated. The in-process in-process inventory quantity data in the assembly process 10 detected by the in-process in-process inventory quantity detection unit 132 is transferred to the assembly shipment date / time quantity probability density calculation unit 115. The in-process inventory quantity data in the assembly process 10 is temporarily stored in the database and then transferred from the database to the assembly / shipping date / time quantity probability density calculation unit 115. Good.

(Description of the inter-process data calculation unit in the third embodiment)
The inter-process data calculation unit includes an inter-process lead time detection unit 231 and an inter-process in-process inventory quantity detection unit 232 as parts for collecting various data of the inter-process inventory 20. Further, the inter-process data calculation unit includes an inter-process lead time storage unit 221 as a part for storing various collected data of the inter-process inventory 20. Further, the inter-process data calculation unit includes an inter-process lead time probability density calculation unit 211 and an inter-process shipment date / time quantity probability density calculation unit 215 as parts for calculating necessary data of the inter-process inventory 20.

  Since the inter-process lead time detection unit 231 has been described in the first embodiment, description thereof will not be repeated. The lead time data of the inter-process inventory 20 of each product detected by the inter-process lead time detection unit 231 is accumulated in the inter-process lead time storage unit 221. The accumulated lead time data of the inter-process inventory 20 is used by the inter-process lead time probability density calculation unit 211 and the total lead time calculation unit 413.

  Since the inter-process lead time probability density calculation unit 211 has been described in the first embodiment, description thereof will not be repeated. The lead time probability density data of the inter-process inventory 20 calculated by the inter-process lead time probability density calculating unit 211 is an assembly shipping date / time quantity probability density calculating unit 115 and an inter-process shipping date / time quantity probability density calculating unit 215 to be described later. Forwarded to

  Since the in-process in-process inventory quantity detection unit 232 has been described in the first embodiment, description thereof will not be repeated. The inventory quantity data of the inter-process inventory 20 detected by the inter-process in-process inventory quantity detection unit 232 is transferred to the inter-process shipment date / time quantity probability density calculation unit 215 described later. However, the present invention is not limited to this, and the inventory quantity data of the inter-process inventory 20 may be temporarily stored in the database and then transferred from the database to the inter-process shipment date / time quantity probability density calculation unit 215. .

(Description of Inspection Process Data Calculation Unit in Third Embodiment)
The inspection process data calculation unit includes an inspection lead time detection unit 331, an in-process in-process inventory quantity detection unit 332, and an inspection non-defective product rate detection unit 333 as parts for collecting various data of the inspection process 30.

  The inspection process data calculation unit includes an inspection lead time storage unit 321 and an inspection acceptable product rate storage unit 322 as parts for storing various data collected in the inspection process 30. Further, the inspection process data calculation unit includes an inspection lead time probability density calculation unit 311, an inspection non-defective product rate probability density calculation unit 313, and an inspection shipment date / time quantity probability density calculation unit as parts for calculating necessary data of the inspection process 30. 315.

  Since inspection lead time detection unit 331 has been described in the first embodiment, description thereof will not be repeated. The inspection lead time data of each product detected by the inspection lead time detection unit 331 is accumulated in the inspection lead time storage unit 321. The accumulated lead time data of the inspection process 30 is used by an inspection lead time probability density calculation unit 311 and an overall lead time calculation unit 413 described later.

  Since inspection lead time probability density calculation unit 311 has been described in the first embodiment, description thereof will not be repeated. The lead time probability density data of the inspection process 30 calculated by the inspection lead time probability density calculating unit 311 includes an assembly shipping date / time quantity probability density calculating unit 115, an inter-process shipping date / time quantity probability density calculating unit 215, and an inspection. It is transferred to the shipment date / time quantity probability density calculation unit 315.

  Note that the probability density data of the calculated lead time of the inspection process 30 is temporarily stored in the database, and then the assembly shipment date / time quantity probability density calculation unit 115 and the inter-process shipment date / time quantity are calculated. It may be transferred to the probability density calculation unit 215 and the inspection shipment date / time quantity probability density calculation unit 315.

  The inspection acceptable product rate detection unit 333 has been described in the second embodiment, and therefore description thereof will not be repeated. The non-defective product rate data in the inspection process 30 detected by the test non-defective product rate detection unit 333 is accumulated in the test non-defective product rate storage unit 322. The accumulated non-defective product rate data of the inspection process 30 is transferred to an inspection non-defective product rate probability density calculating unit 313 and a comprehensive non-defective product rate probability density calculating unit 415 described later.

  The inspection good product rate probability density calculation unit 313 has been described in the second embodiment, and therefore description thereof will not be repeated. The probability density data of the non-defective product rate of the inspection process 30 calculated by the inspection non-defective product rate probability density calculating unit 313 includes an assembly shipping date / time quantity probability density calculating unit 115, an inter-process shipping date / time quantity probability density calculating unit 215, and an inspection. It is transferred to the shipment date / time quantity probability density calculation unit 315.

  The probability density data of the non-defective product ratio of the inspection process 30 calculated is not limited to this, but is temporarily stored in the database, and then the assembly shipment date / time quantity probability density calculation unit 115 and the inter-process shipment date / time quantity are stored in the database. It may be transferred to the probability density calculation unit 215 and the inspection shipment date / time quantity probability density calculation unit 315.

  Since the in-process inventory quantity detection unit 332 in the inspection process has been described in the first embodiment, description thereof will not be repeated. The in-process inventory quantity data in the assembly process 30 detected by the in-process in-process inventory quantity detection unit 332 is transferred to an inspection shipment date / time quantity probability density calculation unit 315 to be described later. The in-process inventory quantity data in the inspection process 30 may be temporarily stored in the database and then transferred to the inspection shipment date / time quantity probability density calculation unit 315.

(Description of the finished product data calculation unit in the third embodiment)
The finished product data calculation unit includes a finished product inventory quantity detection unit 431 as a part for collecting various data of the finished product inventory 40. Since the finished product inventory quantity detection unit 431 has been described in the first embodiment, description thereof will not be repeated. The finished product inventory quantity data stored as inventory before shipment detected by the finished product inventory quantity detection unit 431 is transferred to the production input quantity calculation unit 512B and the shortage excess inventory occurrence probability calculation unit 513B.

(Summary of data calculation units between processes and processes in the third embodiment)
The assembly shipment date / time quantity probability density calculation unit 115 receives the in-process inventory quantity data of the assembly process 10 transferred from the in-process in-process inventory quantity detection unit 132 and the assembly lead time probability density calculation unit 111. The lead time probability density data of the assembly process 10, the lead time probability density data of the inter-process inventory 20 transferred from the inter-process lead time probability density calculation unit 211, and the inspection lead time probability density calculation unit 311 The probability density data of the lead time of the inspection process 30 that has been transferred, the probability density data of the non-defective product ratio of the assembly process 10 that has been transferred from the assembly non-defective rate probability density calculation unit 113, and the transfer from the inspection non-defective product rate probability density calculation unit 313 Based on the probability density data of the non-defective product rate in the inspection process 30 that has been performed, the in-process inventory in the assembly process 10 finishes the assembly process 10. And, and calculates the probability density of the inter-process inventory 20 and working with the through by shippable the probability density of the time required until possible delivery quantity in the inspection process 30.

  Table 8 is an example of probability density data of the shipping date and time of the in-process inventory in the assembly process 10. Table 9 is an example of probability density data of the quantity that can be shipped for the in-process inventory in the assembly process 10.

  As shown in Table 8, the assembly / shipping date / time quantity probability density calculation unit 115 expects the probability density of the shipping possible date / time of the in-process inventory in the assembly process 10 every two hours and the maximum quantity that can be shipped for every shipping available date / time. Value. The maximum quantity that can be shipped is the quantity that can be shipped when the yield rate is 100%. In Table 8, as an example, the in-process in-process inventory quantity detection unit 132 indicates the expected quantity that can be shipped at each date and time when the in-process in-process inventory quantity in the assembly process 10 is detected as 500 units. ing.

  For example, first, a combination of the lead time of the assembly process 10, the lead time of the inter-process inventory 20 and the lead time of the inspection process 30 that can be shipped from 8:00 to 10:00 on January 15 is obtained. Then, by multiplying the probability density corresponding to each step of the obtained combination and the lead time between the steps and adding all the obtained combinations, the in-process in-process inventory of the assembly step 10 is 8 on January 15th. The probability density that can be shipped from 0:00 to 10:00 is calculated. The expected value of the quantity that can be shipped from 8:00 to 10:00 on January 15 in the in-process inventory of the assembly process 10 is calculated according to the calculated probability density in the process of the assembly process 10 here. Calculated by multiplying the in-process inventory quantity by 500 units.

  As shown in Table 9, the assembly / shipping date / time quantity probability density calculation unit 115 calculates probability density data of the quantity that can be shipped with respect to the non-defective product rate for every 0.5% of the in-process inventory in the assembly process 10. In Table 9, as an example, the expected quantity that can be shipped at each date and time when the in-process in-process inventory quantity detecting unit 132 detects 200 in-process in-process inventory quantities in the assembly process 10 is shown. ing.

  For example, the shipping date / time of the entire in-process inventory in the assembly process 10 can be calculated by adding the average values of the lead times of the assembly process 10, the interprocess inventory 20, and the inspection process 30. Further, a combination of the non-defective product ratios of the assembling process 10 and the inspecting process 10 in which the overall non-defective product ratio obtained by multiplying the non-defective product ratios of the assembly process 10 and the inspecting process 10 is 99.5% or more and less than 100% is obtained. Then, the probability density of the whole non-defective product rate is calculated by multiplying the probability density of the non-defective product rate of each step of the obtained combination.

  Furthermore, when the overall non-defective product rate is 100%, the entire in-process in-stock quantity in the assembly process 10 is the quantity that can be shipped. Therefore, the quantity that can be shipped when the overall non-defective rate is 99.5% is 199 units by multiplying the quantity that can be shipped when the overall non-defective rate is 100% by 99.5%. And can be calculated. Then, the calculated shipment possible quantity of 199 is set as a shipmentable quantity when the overall non-defective product ratio is 99.5% or more and less than 100%.

  Since the shippable date and time and the non-defective product rate have continuous values, the probability density function is a continuous function. However, it is not always possible to express the function form by a mathematical expression. Further, even when expression by a mathematical expression is possible, it is not always easy to perform numerical analysis using the mathematical expression. For this reason, in this embodiment, the shipping date / time and the probability density of the non-defective product rate are discretely expressed as shown in Table 8 and Table 9, respectively.

  The assembly shipment date / time quantity probability density calculation unit 115 performs the process of the assembly process 10 on the basis of the calculated probability density data of the in-process in-process inventory of the assembly process 10 and the probability density data of the quantity available for shipment. Probability density data of the quantity that can be shipped for each shipping date and time of in-process inventory is calculated. Table 10 is an example of probability density data of the quantity that can be shipped for each shipping date and time of the in-process inventory in the assembly process 10.

  As shown in Table 10, the assembling / shipping date / time quantity probability density calculation unit 115 calculates the shipping possible date / time of the in-process inventory quantity in the assembling process 10 and the probability density of the available quantity for each non-defective product rate in the assembling process 10. Note that the quantity that can be shipped is calculated by multiplying the expected maximum quantity that can be shipped for each available date and time by the non-defective product rate in the assembly process 10.

  Probability density data of in-process in-process inventory in the assembly process 10 calculated by the assembly / shipping date / time quantity probability density calculation unit 115, probability density data of in-process availability quantity, and in-process in-process inventory in the assembly process 10 The probability density data of the quantity that can be shipped for each date and time is transferred to the whole in-process inventory shipment date and time quantity probability density calculation unit 416 described later.

  Note that the present invention is not limited to this, and the probability density data of the shipping possible date / time of the in-process inventory in the assembly process 10, the probability density data of the shipping quantity available, and the shipping date / time of the in-process inventory of the assembly process 10 The probability density data of the quantity that can be shipped may be temporarily stored in the database and then transferred from the database to the whole in-process inventory shipment date / time quantity probability density calculation unit 416.

  The inter-process shipment date / time quantity probability density calculation unit 215 transfers the in-process in-process inventory quantity data of the inter-process inventory 20 transferred from the inter-process in-process inventory quantity detection unit 232 and the inter-process lead time probability density calculation unit 211. From the inter-process inventory 20 lead time probability density data, the inspection lead time probability density data of the inspection process 30 transferred from the inspection lead time probability density calculation unit 311, and the inspection good product rate probability density calculation unit 313 Based on the probability density data of the non-defective product rate of the inspection process 30 that has been transferred, the inter-process inventory 20 is converted into the inter-process inventory 20 and the inspection process 30 in the same format as the data shown in Table 8 and Table 9. The probability density of time and the probability density of the quantity that can be shipped are calculated until it can be shipped through the work.

  Then, the inter-process shipping date / time quantity probability density calculation unit 215 performs the data shown in Table 10 based on the calculated probability density data of the shipping date / time of the inter-process inventory 20 and the probability density data of the shipping quantity. The probability density data of the quantity that can be shipped for each shipping date of the inter-process inventory 20 is calculated in the same format as above.

  The inter-process shipping date / time quantity probability density calculation unit 215 calculates the probability density data of the shipping date / time of the inter-process inventory 20, the probability density data of the shipping quantity and the probability density data of the shipping quantity for each shipping date / time. It is transferred to a work-in-progress inventory whole shipment date / time quantity probability density calculation unit 416 described later.

  However, the present invention is not limited to this, and the calculated probability density data of the shipping date / time of the inter-process inventory 20, the probability density data of the shipping available quantity, and the probability density data of the shipping quantity for each shipping date / time of the inter-process inventory 20. May be once stored in the database and then transferred from the database to the entire in-process inventory shipment date / time quantity probability density calculation unit 416.

  The inspection shipment date / time quantity probability density calculation unit 315 receives the in-process inventory quantity data of the inspection process 30 transferred from the in-process in-process inventory quantity detection unit 332 and the inspection lead time probability density calculation unit 311. Based on the probability density data of the lead time of the inspection process 30 and the probability density data of the non-defective product rate of the inspection process 30 transferred from the inspection non-defective product probability density calculation unit 313, the in-process inventory of the inspection process 30 is determined. The probability density of the time required to be able to ship through the work in the inspection process 30 and the probability density of the quantity that can be shipped are calculated.

  Then, the inspection shipment date / time quantity probability density calculation unit 315 is the same as the data shown in Table 10 based on the calculated probability density data of the available shipment date / time of the inspection process 30 and the probability density data of the quantity available for shipment. The probability density data of the quantity that can be shipped for each shipment date and time of the in-process inventory in the inspection process 30 is calculated in a simple format.

  Probability density data of in-process in-process inventory of inspection process 30 calculated by inspection shipment date / time quantity probability density calculation unit 315, probability density data of deliverable quantity, and probability density of deliverable quantity for each shipment date / time The data is transferred to an in-process inventory whole shipment date / time quantity probability density calculation unit 416, which will be described later.

  Note that the present invention is not limited thereto, and the calculated probability density data of the in-process in-process inventory in the inspected process 30 that can be shipped, the probability density data of the in-process in-process inventory, and the shipping date and time of the in-process in-process inventory of the inspecting process 30 The probability density data of the quantity that can be shipped may be temporarily stored in the database and then transferred from the database to the whole in-process inventory shipment date / time quantity probability density calculation unit 416.

  In this way, each process and the data calculation unit between processes, the in-process inventory quantity between the processes or processes, the probability density of the lead time calculated from the past lead times between the processes or processes, the past of the processes Based on the probability density of the non-defective product rate calculated from the non-defective product rate, the probability density of the lead time calculated from the past lead time of the post-process, and the probability density of the non-defective product rate calculated from the past non-defective product rate of the post-process The probability density data of the shipping date and time of the in-process inventory between each process and the process, the probability density data of the shipping available quantity, and the probability density data of the shipping quantity for each shipping date and time are calculated.

(Description of the shortage and excess inventory prediction unit in the third embodiment)
The shortage and excess inventory prediction unit includes an overall lead time calculation unit 413, an overall non-defective product rate probability density calculation unit 415, an in-process inventory shipment date / time quantity probability density calculation unit 416, a shipment request quantity storage unit 521, and a shipment request quantity prediction. A value calculation unit 511, a production input quantity calculation unit 512B, and a shortage excess inventory occurrence probability calculation unit 513B are included.

  Since the total lead time calculation unit 413 has been described in the second embodiment, description thereof will not be repeated. The total lead time data until the newly produced input product calculated by the total lead time calculating unit 413 can be shipped is transferred to the shortage excess inventory occurrence probability calculating unit 513B described later.

  However, the present invention is not limited to this, and the calculated total lead time data may be once stored in the database and then transferred from the database to the shortage excess inventory occurrence probability calculation unit 513B.

  Since the total non-defective rate probability density calculation unit 415 has been described in the second embodiment, description thereof will not be repeated. The overall quality rate probability density data of the entire factory calculated by the overall quality rate probability density calculation unit 415 is transferred to the shortage excess inventory occurrence probability calculation unit 513B described later.

  However, the present invention is not limited to this, and the calculated probability density data of the total acceptable product rate may be temporarily stored in the database and then transferred from the database to the insufficient excess inventory occurrence probability calculating unit 513B.

  The in-process inventory whole shipment date / time quantity probability density calculation unit 416 includes the in-process in-process in-process inventory date / time probability density data and the shipment possibility quantity probability density data transferred from the assembly / shipping date / time quantity probability density calculation unit 115. The inter-process shipment date / time quantity probability density calculation unit 215 and the inter-process inventory 20 transferable date / time probability density data and the shipmentable quantity probability density data, and the inspection / shipping date / time quantity probability density calculation unit 315 The probability density of the shippable date and time and the probability density of the quantity that can be shipped for the entire in-process inventory in the factory are calculated from the shippable date / time probability density data and the shippable quantity probability density data of the in-process stock in process 30.

  Then, the in-process inventory whole shipment date / time quantity probability density calculation unit 416 calculates probability density data of the available quantity for each shipment date / time of the entire in-process inventory in the factory in the same format as the data shown in Table 10.

  The shipment date / time probability density data, shipment quantity probability density data, and shipment quantity probability density data of the entire in-process inventory of the factory calculated by the whole work in progress shipment date / time quantity probability density calculation unit 416 will be described later. It is transferred to the production input quantity calculation unit 512B.

  However, the present invention is not limited to this, and the calculated shipping date / time probability density data, shipping quantity probability density data of the entire in-process inventory of the factory, and probability density data of the quantity available for each shipping date / time are temporarily stored in the database. After that, it may be transferred from the database to the production input quantity calculation unit 512B.

  Since the shipping request quantity storage unit 521 and the shipping request quantity predicted value calculation unit 511 have been described in the first embodiment, description thereof will not be repeated. The probability density data of the shipping requested quantity predicted value calculated by the shipping requested quantity predicted value calculating unit 511 is transferred to the production input quantity calculating unit 512B and the shortage excess inventory occurrence probability calculating unit 513B.

  The production input quantity calculation unit 512B includes the shipment request quantity prediction value transferred from the shipment request quantity prediction value calculation unit 511, the factory in-process inventory date / time quantity probability density calculation unit 416, and the processes in the factory and between the processes. The production input quantity is calculated from the available quantity for each shipment date and time of the entire in-process inventory and the finished product inventory quantity transferred from the finished product inventory quantity detection unit 431.

  For example, the production input quantity calculation unit 512B uses the representative value of the shipment request quantity predicted value transferred from the shipment request quantity predicted value calculation unit 511 to calculate each of the factory transferred from the in-process inventory whole shipment date / time quantity probability density calculation unit 416. The quantity obtained by subtracting the expected value of the quantity that can be shipped for each shipment date and time of the entire process and the in-process inventory between each process and the finished product inventory quantity transferred from the finished product inventory quantity detection unit 431, Calculated as production input quantity. Then, the production input quantity calculation unit 512B transfers the calculated production input quantity to the shortage excess inventory occurrence probability calculation unit 513B.

  The shortage excess inventory occurrence probability calculation unit 513B includes the production input quantity transferred from the production input quantity calculation unit 512B and the total lead time data until the new production input transferred from the total lead time calculation unit 413 can be shipped. And the probability density data of the overall non-defective product rate transferred from the general non-defective rate probability density calculating unit 415, the new input product when the quantity calculated by the production input quantity calculating unit 512B is put into production. The probability density of the shippable date and time and the probability density of the shipable quantity are calculated.

  Then, the shortage / overstock occurrence probability calculation unit 513B calculates probability density data of the quantity that can be shipped for each shipping date and time of the newly produced input product in the same format as the data shown as Table 10.

  Further, the shortage excess inventory occurrence probability calculation unit 513B calculates the probability density data of the calculated shipment quantity of newly produced input goods and the entire in-process inventory of the factory transferred from the entire in-process inventory shipment date / time probability probability density calculation unit 416. Probability density data of quantity available for each shipment date and time, probability density data of shipment request quantity predicted value transferred from shipment request quantity predicted value calculation unit 511, and finished product inventory transferred from finished product inventory quantity detection unit 431 From the quantity data, the probability density that the shortage or excess inventory occurs is calculated.

  The calculation of the shortage and excess inventory occurrence probability is performed by the method described with reference to FIG. The shortage excess inventory occurrence probability density calculated by the shortage excess inventory occurrence probability calculation unit 513B is transferred to the shortage excess inventory loss expected value calculation unit 514B and the sellable quantity expected value calculation unit 515B.

(Explanation of Expected Profit Value Calculation Unit in Third Embodiment)
The expected profit value calculation unit includes a shortage excess inventory loss expected value calculation unit 514B, a sellable volume expected value calculation unit 515B, a sales profit expected value calculation unit 516B, and an overall profit expected value calculation unit 517B.

  The shortage excess inventory loss expected value calculation unit 514B calculates the expected value of the loss cost associated with the shortage excess inventory occurrence based on the shortage excess inventory occurrence probability density transferred from the shortage excess inventory occurrence probability calculation unit 513B.

Calculation of the expected value of the shortage and excess inventory occurrence quantity is performed by the method described with reference to FIG. The occurrence probability of the shortage inventory is obtained by calculating ΣΣ {P ₁ (N _i ) × P ₂ (M _j )} for all i, j where i> j. The occurrence probability of excess inventory is obtained by calculating ΣΣ {P ₁ (N _i ) × P ₂ (M _j )} for all i, j where i <j.

The expected value of the shortage of inventory is obtained by calculating ΣΣ {(N _i −M _j ) × P ₁ (N _i ) × P ₂ (M _j )} for all i and j where i> j. It is done.

  The expected value of the loss cost associated with the shortage inventory is calculated by multiplying the expected value of the shortage inventory occurrence quantity thus obtained by the sum of the sales opportunity loss cost per vehicle and the credit compensation cost per vehicle.

Further, the expected value of the excess inventory generation quantity is to calculate ΣΣ {(M _j −N _i ) × P ₁ (N _i ) × P ₂ (M _j )} for all i, j where i <j. Is obtained.

  By multiplying the expected value of the excess inventory generated in this way by the sum of the inventory interest burden per unit, inventory management cost, warehouse cost, and unsold product risk factor cost, the expected loss cost associated with excess inventory A value is calculated.

  The expected value data of the shortage inventory and the loss cost associated with the occurrence of excess inventory obtained by the shortage excess inventory loss expectation value calculation unit 514B is transferred to the total profit expectation value calculation unit 517B.

  Based on the shortage inventory and excess inventory occurrence probability data transferred from the shortage and excess inventory occurrence probability calculation unit 513B, the saleable quantity expectation value calculation unit 515B calculates the expected quantity available for sale at the target date and time.

An example of a method for calculating the expected quantity that can be sold will be described with reference to FIG. Since N _i ≦ M _j , the sellable quantity is N _i , ΣΣ {N _i × P ₁ (N _i ) × P ₂ (M _j )} is set for all i and j where i ≦ j. calculate.

On the other hand, since the sellable quantity when N _i > M _j is M _j , ΣΣ {M _j × P ₁ (N _i ) × P ₂ (M _j ) for all i and j where i> j. } Is calculated. Then, ΣΣ {min (N _i , M _j ) × P ₁ (N _i ) × P ₂ (M _j )}, which is the sum of these calculated values, becomes the expected quantity that can be sold.

  The expected saleable quantity data calculated by the expected saleable quantity calculation unit 515B is transferred to the expected sales profit value calculation part 516B.

  The sales profit expected value calculation unit 516B calculates the sales profit expected value by multiplying the sales possible quantity expected value transferred from the sales possible quantity expected value calculation unit 515B by the sales profit per unit. The sales profit expected value data calculated by the sales profit expected value calculation unit 516B is transferred to the total profit expected value calculation unit 517B.

  The expected total profit value calculation unit 517B calculates the loss cost associated with the shortage of excess inventory and the excess inventory generated from the expected shortage excess inventory loss calculation unit 514B from the expected sales profit value transferred from the expected sales profit value calculation unit 516B. The expected value of total profit is calculated by subtracting the expected value. The calculated total profit is presented to the user together with each parameter of the calculation process such as the production input quantity.

  With the above production planning support device 1B, the probability of occurrence of shortage and excess inventory can be determined from the probability distribution of the production lead time, the probability distribution of the non-defective product rate, and the probability distribution of the predicted quantity requested for shipment at the product manufacturing plant. And the expected value of the total profit can be calculated.

(Description of the flow of processing in the production planning support device 1B in the third embodiment)
Next, a specific process of the production plan planning support apparatus 1B in the third embodiment described above will be described with reference to the flowchart shown in FIG. FIG. 13 is a flowchart showing the flow of a production plan planning support process executed by the production plan planning support apparatus 1B in the third embodiment.

  First, the production planning support device 1B executes the process data calculation process C for the inspection process 30 (step S301). FIG. 14 is a flowchart showing the flow of the process-by-process data calculation process C executed as a subroutine of the production plan planning support process in the third embodiment.

  Referring to FIG. 14, here, a case where the process data calculation process C is executed for the inspection process 30 will be described. First, the inspection lead time probability density calculation unit 311 reads the past lead time of the inspection process 30 from the inspection lead time storage unit 321 (step S41).

  The lead time of the inspection process 30 is detected by the inspection lead time detection unit 331 every time the target product passes the inspection process 30. Then, the detected lead time of the inspection process 30 is stored in the inspection lead time storage unit 321 by the inspection lead time detection unit 331.

  Next, the inspection lead time probability density calculation unit 311 calculates the probability density of the lead time of the inspection process 30 based on the past lead time of the inspection process 30 read in step S41 (step S42).

  Next, the inspection good product rate probability density calculation unit 313 reads the past good product rate of the inspection process 30 from the inspection good product rate storage unit 322 (step S43).

  The non-defective product rate in the inspection process 30 is detected by the test non-defective product rate detection unit 333 every time the target product passes the inspection process 30. The detected non-defective product rate in the inspection process 30 is stored in the inspected non-defective product rate storage unit 322 by the test non-defective product rate detecting unit 333.

  Next, the inspection non-defective rate probability density calculation unit 313 calculates the probability density of the non-defective product rate of the inspection process 30 based on the past non-defective product rate of the inspection process 30 read in step S43 (step S44).

  Further, the in-process inventory quantity detection unit 332 in the inspection process detects the in-process inventory quantity in the inspection process 30 (step S45).

  Next, the inspection shipping date / time quantity probability density calculation unit 315 calculates the probability density of the shipping possible date / time after all the processes in-process inventory in the inspection process 30 (step S46). Further, the inspection shipment date / time quantity probability density calculation unit 315 calculates the probability density of the quantity that can be shipped after all the processes in the in-process inventory in the inspection process 30 (step S47).

  Then, the inspection shipping date / time quantity probability density calculation unit 315 calculates the probability density of the quantity that can be shipped for each shipping date / time after all the in-process inventory in the inspection process 30 (step S48). Thereafter, the process returns to the production planning support process.

  Returning to FIG. 13, the production planning support device 1B executes the inter-process data calculation process b for the inter-process inventory 20 (step S302). FIG. 15 is a flowchart showing the flow of the inter-process data calculation process b executed as a subroutine of the production plan planning support process in the third embodiment.

  With reference to FIG. 15, here, a case where the inter-process data calculation process b is executed for the inter-process inventory 20 will be described. First, the inter-process lead time probability density calculation unit 211 reads the past lead time of the inter-process inventory 20 from the inter-process lead time storage unit 221 (step S51).

  The lead time of the inter-process inventory 20 is detected by the inter-process lead time detection unit 231 every time the target product passes through the inter-process inventory 20. The detected lead time of the inter-process inventory 20 is stored in the inter-process lead time storage unit 221 by the inter-process lead time detection unit 231.

  Then, the inter-process lead time probability density calculation unit 211 calculates the probability density of the lead time of the inter-process inventory 20 (step S52). Next, the in-process in-process inventory quantity detection unit 232 detects the inventory quantity of the in-process inventory 20 (step S53).

  Next, the inter-process shipping date / time quantity probability density calculation unit 215 calculates the probability density of the shipping date / time after all the processes of the inter-process inventory 20 (step S54). Further, the inter-process shipment date / time quantity probability density calculation unit 215 calculates the probability density of the quantity that can be shipped after all the processes of the inter-process inventory 20 (step S55).

  Then, the inter-process shipment date / time quantity probability density calculation unit 215 calculates the probability density of the quantity that can be shipped for each shipment date / time after all the processes of the inter-process inventory 20 (step S56). Thereafter, the process returns to the production planning support process.

  Returning to FIG. 13, the production plan planning support apparatus 1 </ b> B executes the process-by-process data calculation process C for the inspection process 30 as in step S <b> 301 (step S <b> 303).

  Next, the in-process inventory total shipment date / time quantity probability density calculation unit 416 outputs the in-process in-process inventory of each of the assembly process 10, the inter-process inventory 20, and the inspection process 30 calculated in the processes from step S 301 to step S 303. From the probability density data of the available date and time and the quantity that can be shipped, the probability density of the available date and time and the probability density of the quantity that can be shipped for the entire in-process inventory in the factory are calculated (step S304).

  Next, the finished product inventory quantity detection unit 431 detects the quantity of the finished product inventory (step S305). Further, the shipment request quantity predicted value calculation unit 511 reads the past shipment request quantity from the shipment request quantity storage unit 521 (step S306). Then, the shipping request quantity predicted value calculation unit 511 calculates the shipping request quantity predicted value based on the accumulated data of the shipping request quantity from the past read in step S306 (step S307).

  Next, the production input quantity calculation unit 512B calculates the expected value of the quantity that can be shipped for each shipping date and time of the entire in-process inventory of the factory calculated in step S304, the predicted shipment requested quantity data calculated in step S307, and step S305. The production input quantity is calculated from the finished product inventory quantity data calculated in (Step S308).

  Next, the comprehensive non-defective rate probability density calculation unit 415 is based on the accumulated data of the non-defective product ratios of the assembly process 10 and the inspection process 30 from the past stored in the assembled non-defective product rate storage unit 122 and the inspection non-defective product rate storage unit 322, respectively. The probability density of the total non-defective product rate is calculated (step S309).

  The total lead time calculation unit 413 includes an assembly lead time storage unit 121, an inter-process lead time storage unit 221, and an assembly lead time storage unit 321 from the past, the assembly process 10 from the past, the inter-process inventory 20, respectively. Then, the probability density of the total lead time is calculated based on the accumulated lead time data of the inspection process 30 (step S310).

  Next, the shortage / overstock occurrence probability calculation unit 513B calculates the probability density data of the total non-defective product rate calculated in step S309, the total lead time data calculated in step S310, and the production input quantity data calculated in step S308. The expected value of the available quantity for each shipment date and time of the entire in-process inventory in the factory calculated in step S304, the predicted shipment quantity predicted value data calculated in step S307, and the finished product inventory quantity calculated in step S305. From the data, the occurrence probability of the shortage inventory and excess inventory is calculated (step S311).

  Based on the occurrence probability data of the shortage inventory and excess inventory calculated in step S311, the shortage and excess inventory loss expected value calculation unit 514B calculates the expected value of the loss cost associated with the shortage inventory and excess inventory (step S312). Further, the expected quantity for sale possible quantity calculation unit 515B calculates the expected value for the quantity available for sale based on the occurrence probability data for the shortage and excess inventory calculated in step S311 (step S313).

  Then, the expected sales profit value calculation unit 516B calculates the expected value of sales profit based on the expected value data of the available quantity calculated in step S313 (step S314).

  Finally, the total profit expected value calculation unit 517B calculates the total profit based on the expected value data of the loss cost accompanying the shortage and excess inventory calculated in step S312 and the expected value data of the sales profit in step S314. An expected value is calculated (step S315).

(Operations and effects of the production planning support device 1B in the third embodiment)
As described above, the production plan planning support device 1B in the third embodiment is a device that supports the planning of a product production plan. The production planning support device 1B includes a deficiency excess inventory prediction unit and a profit expected value calculation unit.

  The shortage and excess inventory forecasting section is based on production variables such as yield rate data for each process, lead time data for each process or process, and in-process inventory quantity data. Calculate the occurrence probability of inventory and excess inventory. The expected profit value calculation unit calculates the expected value of the total profit based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage and excess inventory prediction unit.

  As a result, the production planning support device 1B calculates the expected value of the total profit based on the occurrence probability of the shortage inventory and excess inventory with respect to the current and future production input quantities calculated from the variables relating to production. For this reason, it is possible to support the planning of production plans by estimating the total profit in the event of shortage or excess inventory. As a result, the production planning accuracy can be improved by the production planning support device 1B. Further, the production planning support device 1B can contribute to the production planning considering the management numerical value.

  The variables related to production are past statistical values, current measured values, and future predicted values. The current actual measurement values are the lead time in each process of production detected by the assembly lead time detection unit 131, the inter-process lead time detection unit 231 and the inspection lead time detection unit 331, and the assembly good product rate detection unit 133 and the inspection good product. The non-defective product ratio in each production process detected by the rate detection unit 333, the finished product inventory quantity of products waiting for shipment after production detected by the finished product inventory detection unit 431, and the assembly process Each process detected by the in-process inventory quantity detection unit 132, the inter-process in-process inventory quantity detection unit 232, and the in-process in-process inventory quantity detection unit 332, and the in-process inventory quantity of the product between the processes, the shipping request quantity storage unit 521 The requested shipping quantity stored in

  The past statistical values are obtained from the assembly lead time storage unit 121, the inter-process lead time storage unit 221, and the lead time accumulated in the inspection lead time storage unit 321, respectively. The inter-process lead time probability density calculating unit 211 and the inspection lead time probability density calculating unit 311, the probability density of each process and the lead time between the processes, the assembly non-defective product rate storage unit 122, the inspection good product rate storage unit 322, The probability ratio of the non-defective product rate of each process calculated by the assembly non-defective product rate probability density calculating unit 113 and the inspection non-defective product rate probability density calculating unit 313, and the requested shipping quantity predicted value calculating unit 511, respectively. The probability density of the requested shipping quantity calculated by

  The future predicted value is a predicted value of the requested shipping quantity of the product calculated by the shipping requested quantity predicted value calculation unit 511.

  The production planning support device 1B includes an actual measurement value detection unit that detects a current actual measurement value, an actual measurement value storage unit that stores an actual measurement value detected by the actual measurement value detection unit, and an actual measurement value stored in the actual measurement value storage unit. Select a statistical value calculation unit that calculates the probability density of the actual measurement value as a past statistical value, a predicted value calculation unit that calculates a future predicted value based on the statistical value calculated by the statistical value calculation unit, etc. Including.

  The actual measurement value detection unit includes an assembly non-defective product rate detection unit 133, an assembly lead time detection unit 131, an in-process in-process inventory quantity detection unit 132, an inter-process lead time detection unit 231 and an in-process in-process inventory quantity detection unit. 232, an inspection good product rate detection unit 333, an inspection lead time detection unit 331, an in-process in-process inventory quantity detection unit 332, and a finished product inventory quantity detection unit 431.

  The actual measurement value storage unit includes an assembly good product rate storage unit 122, an assembly lead time storage unit 121, an inter-process lead time storage unit 221, an inspection good product rate storage unit 322, and an inspection lead time storage unit 321.

  As the statistical value calculation unit, an assembly lead time probability density calculation unit 111, an assembly good product rate probability density calculation unit 113, an assembly shipment date / time quantity probability density calculation unit 115, an inter-process lead time probability density calculation unit 211, a process Inter-shipment date / time quantity probability density calculation unit 215, inspection lead time probability density calculation unit 311, inspection non-defective product rate probability density calculation unit 313, inspection / shipping date / time quantity probability density calculation unit 315, total non-defective product rate probability density calculation unit 415 In addition, there is a total lead time calculation unit 413, an in-process inventory whole shipment date / time quantity probability density calculation unit 416, and a shipment request quantity predicted value calculation unit 511.

  As the predicted value calculation unit, there is a shipping request quantity predicted value calculation unit 511. In addition, each probability density calculation unit described above may calculate a probability density of a certain variable and calculate a future predicted value of the certain variable from the probability density.

  As a result, the current actual measurement value relating to production is detected by the production planning support device 1B, the detected actual measurement value is stored, and the probability density of the actual measurement value is determined as a past statistical value based on the stored actual measurement value. A future predicted value is calculated based on the calculated statistical value. For this reason, the expected value of total profit is based on the occurrence probability of shortage and excess inventory for the current and future production inputs calculated from current measured values, past statistical values, and future predicted values. Can be calculated.

  The production planning support device 1B further includes a shortage excess inventory loss expected value calculation unit 514B and a sales profit expected value calculation unit 516B. The shortage excess inventory loss expected value calculation unit 514B and the sales profit expectation value calculation unit 516B have the same effects as the shortage excess inventory loss expectation value calculation unit 514 and the sales profit expectation value calculation unit 516 in the first embodiment. Play.

  As the production planning method supported by the application of the present invention described above, various methods such as a production planning method based on simulation and a production planning method based on a numerical operation algorithm can be adopted.

  In each of the above-described embodiments, the production input quantity calculation units 512, 512A, and 512B respectively transfer the calculated production input quantity as it is to the shortage excess inventory occurrence probability calculation units 513, 513A, and 513B. However, the present invention is not limited to this, and the production input quantity calculation units 512, 512A, and 512B may present the calculated production input quantity to the user so that the user can correct the production input quantity. Good. In addition, the production input quantity calculation units 512, 512A, and 512B present a plurality of production input quantity candidates calculated based on the calculated production input quantity to the user, and the user can produce from among the plurality of candidates. The input quantity may be selectable. Thereby, the expected value of the total profit can be calculated in consideration of the user's experience and intuition.

  In each embodiment described above, the production input quantity calculation units 512, 512A, and 512B calculate one production input quantity. However, the present invention is not limited to this, and the production input quantity calculation units 512, 512A, and 512B may calculate a plurality of production input quantities. For example, a quantity of ± 10% with respect to one calculated production input quantity may be calculated as another production input quantity, and the total profit for each of the plurality of production input quantities may be calculated. Thereby, it is possible to present and present the total profit for a plurality of production input quantities to the user. For this reason, it can further contribute to the production plan taking into consideration the management numerical aspect.

  In each of the above-described embodiments, the allocation quantity allocated from the finished product inventory to the shipment may be automatically determined by a predetermined calculation formula. As a result, it is possible to actively determine the allocation quantity according to the inventory quantity of the finished product inventory. Further, the user may be allowed to input the allocation quantity. Thereby, the allocation quantity can be determined in consideration of the user's experience and intuition.

  In each of the above-described embodiments, a detection unit for detecting data relating to production is provided between each process. However, the present invention is not limited to this, and a data detection unit for production may be provided between each process and between processes. For example, in order to identify the products that have passed between the assembly process 10 and the inter-process inventory 20, between the inter-process inventory 20 and the inspection process 30, and between the inspection process 30 and the finished product inventory 40, respectively. A detecting unit for detecting the time when the specified product passes or a detecting unit for detecting the time may be provided. Thereby, based on the detection value obtained by these detection units, the lead time of the product and the yield rate can be calculated.

  In each of the embodiments described above, the production planning support devices 1, 1A, 1B are each configured by one computer. However, the present invention is not limited to this, and the production planning support devices 1, 1 </ b> A, 1 </ b> B may each be configured by a plurality of computers connected by a network. For example, the functions of the assembly process data calculation unit, the inspection process data calculation unit, the inter-process data calculation unit, the finished product data calculation unit, the deficient excess inventory prediction unit, and the expected profit calculation unit are realized by separate computers. You may be made to do.

  In the above-described embodiments, the production planning support devices 1, 1A, 1B have been described. However, the production planning support method in which the processes shown in FIGS. 4, 5, 9 to 11, and 13 to 15 are executed by a computer are shown in FIGS. 4, 5, 9 to 11, and 13 to 13. The invention can be understood as a production plan planning support program for causing a computer to execute the processing shown in FIG. 15 and a computer-readable recording medium on which the production plan planning support program is recorded.

  The production plan planning support apparatus, the production plan planning support method, the production plan planning support program, and the recording medium according to the present invention exhibit variations in demand, manufacturing lead time, and non-defective product rate when planning a production plan in a product manufacturing factory. Taking this into consideration, it can be used effectively when maximizing the expected profit in terms of management.

  The embodiment disclosed this time is to be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

It is a functional block diagram which shows the outline of the function of the production plan planning assistance apparatus in one Embodiment of this invention. It is a graph for demonstrating the calculation method of the shortage excess inventory occurrence probability performed with the production planning support apparatus in 1st Embodiment. It is a graph for demonstrating the calculation method of the shortage excess inventory generated quantity expected value performed with the production plan planning assistance apparatus in 1st Embodiment. It is a flowchart which shows the flow of the production plan planning support process performed with the production plan planning support apparatus in 1st Embodiment. It is a flowchart which shows the flow of the data calculation process A for every process performed as a subroutine of a production plan planning assistance process in 1st Embodiment. It is a functional block diagram which shows the outline of the function of the production plan planning assistance apparatus in 2nd Embodiment of this invention. It is a 1st graph for demonstrating the calculation method of the shortage excess inventory occurrence probability performed with the production planning support apparatus in 2nd Embodiment. It is a 2nd graph for demonstrating the calculation method of the shortage excess inventory occurrence probability performed with the production planning support apparatus in 2nd Embodiment. It is a flowchart which shows the flow of the production plan planning support process performed with the production plan planning support apparatus in 2nd Embodiment. It is a flowchart which shows the flow of the data calculation process B for every process performed as a subroutine of a production plan planning assistance process in 2nd Embodiment. It is a flowchart which shows the flow of the data calculation process a between processes performed as a subroutine of a production plan planning assistance process in 2nd Embodiment. It is a functional block diagram which shows the outline of the function of the production plan planning assistance apparatus in 3rd Embodiment of this invention. It is a flowchart which shows the flow of the production plan planning support process performed with the production plan planning support apparatus in 3rd Embodiment. It is a flowchart which shows the flow of the data calculation process C for every process performed as a subroutine of a production plan planning assistance process in 3rd Embodiment. It is a flowchart which shows the flow of the data calculation process b between processes performed as a subroutine of a production plan planning assistance process in 3rd Embodiment.

Explanation of symbols

  1, 1A, 1B Production planning support device, 10 assembly process, 20 inter-process inventory, 30 inspection process, 40 finished product inventory, 111 assembly lead time probability density calculation unit, 112 assembly shipping possible date / time probability density calculation unit, 113 assembly Non-defective product rate probability density calculation unit, 114 Assembling / shipping quantity probability density calculation unit, 115 Assembly shipping date / time quantity probability density calculation unit, 121 Assembly lead time storage unit, 122 Assembly non-defective product rate storage unit, 131 Assembly lead time detection unit, 132 Assembly In-process in-process inventory quantity detection unit, 133 Assembling non-defective product rate detection unit, 211 Inter-process lead time probability density calculation unit, 212 Inter-process shipable date / time probability density calculation unit, 213 Inter-process shipable quantity probability density calculation unit, 215 Between processes Shipment date / time quantity probability density calculation unit, 221 inter-process lead time storage unit, 231 inter-process lead tie Detection unit, 232 In-process in-process inventory quantity detection unit, 311 Inspection lead time probability density calculation unit, 312 Inspection shipment possible date / time probability density calculation unit, 313 Inspection good product rate probability density calculation unit, 314 Inspection shipment possible quantity probability density calculation unit, 315 Inspection shipment date / time quantity probability density calculation unit, 321 Inspection lead time storage unit, 322 Inspection good product rate storage unit, 331 Inspection lead time detection unit, 332 In-process in-process inventory quantity detection unit, 333 Inspection good product rate detection unit, 411 General Lead time probability density calculating unit, 412 Probability density calculating unit for total shipment of in-process inventory, 413 Total lead time calculating unit, 414 Total probability of shipping in-process inventory quantity probability density calculating unit, 415 Total non-defective product rate probability density calculating unit, 416 In-process inventory Total shipment date / time quantity probability density calculation unit, 431 finished product inventory quantity detection unit, 511 Shipment request quantity predicted value calculation unit, 512, 512A, 512B Production input quantity calculation unit, 513, 513A, 513B Shortage excess inventory occurrence probability calculation unit, 514, 514A, 514B Shortage excess inventory loss expected value calculation unit, 515, 515A, 515B Expected quantity available for sale calculation unit, 516, 516A, 516B Expected sales profit calculation part, 517, 517A, 517B Expected total profit calculation part, 521 Shipment requested quantity storage part.

Claims (14)

A production planning support device for supporting product production planning,
A shortage / overstock occurrence probability calculation unit that calculates a probability of occurrence of shortage and excess inventory with respect to current and future production input quantities from variables related to production;
A production planning support device, comprising: an expected profit value calculation unit that calculates an expected value of total profit based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage and excess inventory occurrence probability calculation unit.   The production variable planning support apparatus according to claim 1, wherein the variables are past statistical values, current measured values, and future predicted values.   The production planning support device according to claim 2, further comprising an actual value detection unit that detects the current actual value.   The current actual measurement value is a shipment-completed inventory quantity of the product waiting for shipment after the production is completed, and an in-process inventory quantity of the product between each process and each process. Production planning support equipment. An actual value storage unit for storing the actual value detected by the actual value detection unit;
The production planning support device according to claim 3, further comprising a statistical value calculation unit that calculates a probability density of the actual measurement value as the past statistical value based on the actual measurement value stored in the actual measurement value storage unit. . The current actual measurement value is a lead time of the product between each step of the production and each step,
The production planning support apparatus according to claim 5, wherein the past statistical value is a probability density of the lead time. The current actual measurement value is a production non-defective rate of the product in each step of the production,
The production planning support apparatus according to claim 5, wherein the past statistical value is a probability density of the non-defective product rate.   The production plan planning support apparatus according to claim 5, further comprising a predicted value calculation unit that calculates the future predicted value based on the statistical value calculated by the statistical value calculation unit. The current actual measured value is a requested shipping quantity of the product,
The past statistical value is a probability density of a requested shipment quantity of the product,
9. The production planning support apparatus according to claim 8, wherein the predicted future value is a predicted shipment quantity predicted value of the product. A shortage excess inventory loss expected value calculation unit that calculates an expected loss value based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage excess inventory occurrence probability calculation unit;
A sales profit expectation value calculation unit that calculates an expected value of sales profit based on the shortage inventory and the occurrence probability of excess inventory calculated by the shortage excess inventory occurrence probability calculation unit;
The expected profit value calculation unit is an expected value of profit obtained by subtracting an expected value of loss calculated by the shortage excess inventory loss expected value calculation unit from an expected value of sales profit calculated by the expected sales profit value calculation unit. The production planning support apparatus according to any one of claims 1 to 9, wherein the value is calculated as an expected value of the total profit. A production planning support device for supporting product production planning,
A lead time detection unit for detecting the lead time of the product between each step of the production and each step;
A lead time storage unit for storing the lead time detected by the lead time detection unit;
A lead time probability density calculating unit that calculates a probability density of the lead time based on the lead time stored in the lead time storage unit;
A non-defective product rate detection unit for detecting a non-defective product rate of the product in each step of the production;
A non-defective product rate storage unit for storing the non-defective product rate detected by the non-defective product rate detection unit;
Based on the manufacturing good product rate stored in the manufacturing good product rate storage unit, the manufacturing good product rate probability density calculating unit that calculates the probability density of the manufacturing good product rate;
A finished product inventory quantity detection unit for detecting a finished product inventory quantity of the product waiting for shipment after completion of the production;
In-process inventory quantity detection unit for detecting the in-process inventory quantity of the product between each process and each process;
A shipment quantity prediction value calculation unit for calculating a shipment quantity prediction value of the product;
The lead time probability density calculated by the lead time probability density calculating unit, the manufacturing non-defective product rate probability density calculated by the non-manufacturing good product rate calculating unit, and the completion of waiting for shipment detected by the finished product inventory quantity detecting unit Shortage of current and future production input quantities based on product inventory quantities, in-process inventory quantities detected by the in-process inventory quantity detection unit, and estimated shipment quantity values calculated by the estimated shipment quantity calculation unit A shortage and excess inventory occurrence probability calculation unit for calculating the occurrence probability of inventory and excess inventory,
A shortage excess inventory loss expected value calculation unit that calculates an expected loss value based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage excess inventory occurrence probability calculation unit;
A sales profit expectation value calculation unit that calculates an expected value of sales profit based on the shortage inventory and the excess inventory occurrence probability calculated by the shortage excess inventory occurrence probability calculation unit;
Total profit expectation value for calculating an expected value of total profit obtained by subtracting an expected value of loss calculated by the shortage excess inventory loss expected value calculation section from an expected value of sales profit calculated by the sales profit expectation value calculation section A production planning support device including a calculation unit. A production planning support method for supporting production planning of a product by a computer,
Calculating the probability of under and over inventory from production variables for current and future production inputs,
Calculating an expected value of total profit based on the calculated occurrence probability of the shortage and excess inventory. A production planning support program for supporting product production planning,
Calculating the probability of under and over inventory from production variables for current and future production inputs,
A production planning support program for causing a computer to execute a step of calculating an expected value of total profit based on the calculated occurrence probability of the shortage and excess inventory.   A computer-readable recording medium on which the production planning support program according to claim 13 is recorded.

Images