JPH1031504A - Automatic preparation method for optimum production schedule and optimum production schedule preparation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten calculation time by using the solution of the minimum cost problem of a network planning method and preparing a production schedule based on an optimum path and the value of a parameter at that time. SOLUTION: A network model where manually related manufacturing elements in manufacture elements are connected ion the from of a network is previously decided. In the network, node groups 20-25 are connected by arcs in accordance with the relation of production processes. When manufacturing relation parameters (paper product manufacturing request quantity and raw material (pulp) compound rate of paper product) are used, the total flow rate of the minimum cost flow problem of the network model is converted into a power rate and it can be displayed by a prescribed system. CPU decides the path (route of node) whose power cost of the network model becomes a minimum, obtains the value of the parameter at that time, obtains the flow rate flowing in the respective arcs at every unit time and decodes the optimum production schedule.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to a production facility,
The present invention relates to an optimal production schedule automatic creation method and an optimal production schedule creation device suitable for a pulp and paper manufacturing facility.

[0002]

2. Description of the Related Art Conventionally, an operation schedule of a production plant, that is, a production schedule, has often been manually created by several experienced planners. However, even an inexperienced person can create a production schedule. A production schedule creation device using a computer has been proposed. When a production schedule is created by a computer, a linear programming method (L
The use of P) is generally well known.

Here, as an example, it is assumed that the power type can be selected from three types in a paper mill and the power cost changes with time, and the total power cost for raw material (pulp) production is used as an objective function. Let us assume a raw material production schedule problem that minimizes this.

[0004] A separately determined paper production (papermaking) plan,
From the mixing ratio of various raw materials (pulp) based on the paper type (grade), the amount of raw material (pulp) used for N days from the present
The value calculated for each unit time of the schedule is the amount of raw material (pulp) taken for each hour.

The pulp storage tank and each raw material in consideration of the good operation of the received power type so as not to hinder the pulp production amount and to minimize the total power cost required for pulp production. A description will be given assuming a problem of creating an optimal production schedule for pulp.

[0006] The power used for production is classified into private power and purchased power according to the type of power. The private power is assumed to be thermal power and hydropower with different generation unit prices, and purchased power is night (Night) band, peak (Peak) The unit price differs depending on the reception time zone such as obi, off peak (Off Peak) zone, and this price difference,
By making the best use of the storage tank (cushion chest) of each raw material (pulp) manufacturing facility, producing and storing as much raw material (pulp) as possible during hours when the unit price of purchased power is low. In addition, in the time zone when the stock volume of the storage tank is low, the raw material (pulp) is always
Production schedules must be created so that there is no excess or deficit.

[0007] To summarize the problems exemplified here,
It looks like this: << Parameters (variables) of constraints >> Raw material (pulp) production amount → Pn, t (Ton / Hr.) ・ Cushion chest raw material storage amount of each raw material → Cn, t (Ton) ・ Electric power consumption → Ee, t (Kw)-Production demand and production time for each grade of paper product in the final (papermaking) process. (Required paper product production) → PPg, m, t
(Ton / Hr.)

[0008] Raw material (pulp) blending ratio of paper products of each grade → Hg, n, m (%) [* The blending ratio is different for each paper machine even for the same grade. ] ・ Electricity unit price → ECe, t (yen / kw ・ H) ・ Emission unit of raw material (pulp) production → EUn (kw ・ H / Ton) ・ Plan to stop each production process. [Raw material (pulp) production equipment, final process (papermaking) equipment]

(Description of Subscripts) n: Type of Raw Material (Pulp) t: Time [Number of (Schedule Time / Calculation Unit Time) ...] e: Type of Electric Power Purchased Power (P-band, OP-band ...) N-band, etc.) Thermal power (private power generation) Hydropower (private power generation) g: Type of paper product (grade) m: Type of paper machine

<< Variables of Upper / Lower Limit Constraints >> The amount of each raw material (pulp) that can be produced, such as KP, TMP1, TMP2, GP, DIP ... → Pn, t Max., Pn, t Min. (Ton / Hr.) ・ Limit the storage capacity of cushion chest of each raw material. → CnMax, CnMin (Ton) ・ Suppliable amount of electric power [in-house power generation (thermal, hydro), purchased power (P-band, OP-band, N-band, etc.)]. → Ee, t Max., Ee, t Min. (Kw) ・ Limited capacity of transfer equipment such as pumps in the manufacturing process.

<< Objective function >> Assuming that total electric power cost → Z (yen), the optimal production schedule of each raw material (pulp), the optimal use schedule of each electric power, and the optimal operation of the storage tank so as to minimize the total electric power cost. Find a pattern.

In the linear programming (LP), which is one of the mathematical programming methods often used when solving such a problem using a computer, it is expressed mathematically as follows. (Constraints)

[Equation 1] Pn, t Min. ≤ Pn, t ≤ Pn, t Max.

[Equation 2] Ee, t min. ≦ Ee, t ≦ Ee, t max.

[0013]

(Equation 3)

[Cn Min. ≤ Cn, t ≤ Cn Max.)

(Equation 5)

(Objective function)

(Equation 6)

When the linear programming method is applied to a production schedule creating apparatus, the manufacturing process from supply of raw materials to obtaining a final product is generally controlled by the above-mentioned various parameters (raw material production, power consumption, etc.). ) To represent a linear polynomial. Then, based on the algorithm of the linear programming, the solution of the parameter that minimizes the objective function; Z is calculated by the computer.

[0016]

In the conventional linear programming, the calculation time (amount) required to find an optimal solution is obtained.
Is approximately the cube of the “number of constraint expressions” and the “unknown number [(P
n, t), (En, t), (Cn, t)] and the storage capacity required for the computer is approximately "the number of constraint equations"
It is empirically known that it increases in proportion to the product of the square of and the number of unknowns.

In an actual production process, if the number of parameters is large and the time length of the plan is increased or the unit time of the plan is shortened, it is necessary to create a large-scale and detailed plan. The required computation time and required storage memory increase significantly. For this reason, the conventional apparatus requires a computer having a large data storage capacity, and requires a long calculation time.

It is an object of the present invention to provide an automatic production schedule automatic production method and an optimal production schedule production apparatus which can reduce the calculation time based on a model completely different from the linear programming method.

[0019]

In order to achieve the above object, according to the first aspect of the present invention, a plurality of manufacturing elements which affect manufacturing are quantified as parameters for each type, and the evaluation is determined by a combination of the parameters. In an automatic production schedule automatic creation method for creating a production schedule by determining a parameter value at which a criterion is optimized by a computer, a network in which mutually related manufacturing elements among the plurality of manufacturing elements are connected in a network form Assuming a model, on the network model, the solution of the minimum cost flow problem of the network planning method is used to obtain the path at which the evaluation criterion is optimal and the value of the parameter at that time. In order to solve the problem, the value of the parameter corresponding to the plurality of manufacturing elements is set to the value of the same kind of parameter. In terms, to obtain the value of a parameter of said optimum The path and at that time every unit time, based on the value of the acquired path and parameters, characterized in that to create the production schedule.

According to a second aspect of the present invention, in the automatic production schedule creating method according to the first aspect, the evaluation criterion has a minimum cost.

According to a third aspect of the present invention, in the method for creating an optimum production schedule according to the first aspect, the production is the production of paper pulp, and a plurality of types of raw materials (pulp) are prepared in the production of the paper pulp. The raw material is blended to produce a plurality of types of products (paper), and the raw material can be stored in a storage tank for each type, and in a manufacturing process in which a plurality of types of electric power having different power unit prices are selectively supplied. , The production time of the product to be produced and the production amount are given as input data, the evaluation criterion is the minimum power cost, the parameters of the same type are calculated as the power consumption, and the raw material production and the power consumption per unit time are calculated. And the amount of storage in the storage tank is obtained to create an optimal production schedule.

According to a fourth aspect of the present invention, a plurality of manufacturing elements which affect the manufacturing are quantified as parameters for each type, and a computer determines a parameter value at which an evaluation criterion determined by a combination of the parameters is optimal. In the optimal production schedule creating apparatus for creating a production schedule, a network model in which mutually related manufacturing elements among the plurality of manufacturing elements are connected in a network form is determined in advance, and the evaluation criterion is defined on the network model. In order to obtain the optimal path and the value of the parameter at that time, the solution of the minimum cost flow problem of the network programming is used, and in order to solve the minimum cost flow problem, By converting the value of the parameter to the value of the same type of parameter, A first information processing means for obtaining a value of Kino parameter, based on the value of the obtained path and parameters, characterized in that comprising a second information processing means for creating a production schedule.

According to a fifth aspect of the present invention, in the optimum production schedule creating apparatus according to the fourth aspect, the structure of the network model can be variably set.

[0024]

Embodiments of the present invention will be described below in detail with reference to the drawings.

In general, the problem of optimization of evaluation criteria (for example, minimum cost) when there are a plurality of parameters (variables) includes:
Linear programming is used, and the least cost flow problem of network programming is used for problems with one parameter (variable). Therefore, linear programming is conventionally used for the problem with several types of parameters as described in the section of the prior art. However, in the present invention, several types of parameters are converted into one parameter, and the production process is reduced. The first feature of the present invention resides in devising modeling as a network so that the solution of the minimum cost flow problem can be applied to such a problem. Also, when applying the minimum cost flow problem of network planning, a unit time is set,
The second feature is that the optimal solution for each unit time in the manufacturing process model is obtained.

Further, when determining the production schedule, there is a problem as to what is used as a reference for evaluation. In the present invention, as an evaluation criterion, the manufacturing cost is minimized.
However, the manufacturing cost includes the cost of purchasing raw materials, the cost of power, the cost of human rights, and the like. The main thing that fluctuates over a short period of time is the cost of power. Therefore, in the present embodiment, attention is paid to the fact that, in the paper mill, the power cost accounts for a relatively large proportion of the manufacturing cost, and there is a price difference depending on the type of power, and the price fluctuates depending on the time. Third, the use of electricity cost as a schedule evaluation criterion
There is a feature. Further, when converting a plurality of parameters into one parameter, there is a fourth feature in that the parameter to be converted is an electric energy.

In a production factory, it is required to produce products efficiently and at a low cost by making the most of the capacity of the production equipment owned. Therefore, in a complicated production process, if a well-developed and optimally-oriented production plan can be automatically created in a short time in consideration of various conditions, it will be very advantageous in production activities.

In the manufacturing process of a paper mill or a chemical mill,
In many cases, products are made by mixing (blend) a plurality of ingredients, and such equipment generally has a plurality of storage tanks (cushion chests) capable of making and storing the ingredients,
This is often used as a buffer facility for storage for stable operation.

On the other hand, the type of electric power for producing the raw material can be selected, and each electric power cost often changes over time. In such a case, it is necessary to solve the scheduling problem of how to select the electric power over time and operate each raw material facility is the most efficient operation. When considering the various conditions and the like, finding an optimal solution that minimizes the power cost requires considerably complicated thinking. Here, in a production process having such characteristics, the production equipment is used effectively, and the production execution schedule that minimizes the total power cost for manufacturing this product as an objective function is automatically set. Use mathematical modeling techniques to create.

More specifically, each raw material production process is designed so that there is no excess or deficiency in the production plan of the final product that is determined to be manufactured in advance, and that the total power cost for raw material production is minimized. Is a method of modeling a mathematical model that automatically calculates the production schedule in units of hours using a computer and a solution to the least cost flow problem of network programming, which is a method of mathematical programming, and its solution.

This method is applicable to any type of plant as long as a plurality of types of parameters are related to each other and have a form capable of being converted to the same type of parameter. It is also applicable to:
Although the calculation scale greatly depends on the computer performance, for example, a large-scale “production schedule” having a span of about 1 to 7 days at intervals of 10 to 120 minutes can be created in a considerably short time. In general, it is rare that the optimal production schedule created automatically can be transferred to the execution level as it is at the production site, but actual operation by partially modifying the human system based on the automatically created schedule Can also be used. Further, even when there is a sudden production change, it is possible to change a part of conditions and easily execute recalculation to quickly create a new production schedule.

An outline of the minimum cost flow problem of network programming is as follows. In the network planning method, as shown in FIG. 1, the state of the analysis target is defined as "node: point" and "arc:
It is represented by a "graph" connected by representing it with an "arc". This network-like "graph" has "source: start point" and "sink: end point", and thinks that an object flows from "source" to "sink" via "arc" and "node".

At this time, if the "cost" flowing through the "arc" is determined for each "arc", the problem of finding a route that can reach the "sink" from the "source" at the "minimum cost" is as follows. Called the shortest problem.

[0034] In addition, if the "maximum capacity" and, in some cases, the "minimum capacity" condition are set for each arc, and assuming that an object flows from the "source" to the "sink," The problem of finding the maximum flow rate, that is, whether the maximum flow rate can be flowed, is called the "maximum flow problem".

The "minimum cost flow problem" is defined for each "arc"
On the other hand, if the "maximum capacity" and "cost" flowing through the "arc" and the "minimum capacity" condition are specified as an applied problem, the desired total flow rate from the "source" to the "sink" This is a problem of finding the optimal flow as to how much flow can be allocated to which route and flow can be realized at "minimum cost" when trying to flow the entire amount.

References regarding network planning include “Network Planning” by Takashi Kobayashi, Baifukan 198.
4. Please refer to “Network Theory” by Masao Iri and Takashi Kobayashi, Nikka Giren Publisher, 1976 and so on.

FIG. 1 shows an example of a simple "minimum cost flow problem" by reference numeral (a), and its optimal solution is shown by reference numeral (b).
The "minimum cost flow problem" is the "shortest problem" and "maximum flow problem"
It is common to solve the problem by finding the "shortest route" and repeatedly assigning the "maximum flow", which is the flow rate that can flow through that route. Is shown in When the following processing is performed on the network indicated by the symbol (a) in FIG. 1 based on this solution, an optimal solution is obtained.

(1) Find the shortest path of →. [→
→→] (2) Find and assign the maximum flow rate that can flow to →→→. [Min. (60,20,40) = 20] (3) Modify the maximum flow rate of the arc. [(→,
→, →) = (60-20, 20-20, 40-20)] (4) Find the second shortest path of →. [→→
→]

(5) Find and assign the maximum flow rate that can flow in →→→. [Min. (60-20, 40, 30) = 30] (6) Correct the maximum current of the arc. [(→, →
, →) = (60-20-30, 40-30, 30-30)] (7) Find the third shortest path of →. [→→
→→] (8) Find and assign the maximum flow that can flow to →→→→. [Min. (60-20-30, 50,50,70) = 10]

(9) The maximum flow of the arc is corrected. [(→, →, →, →) = (60-20-30-1
0, 50-10, 50-10, 70-10)] Find the fourth shortest path of (10) →. :: (11) Hereinafter, the same operation is repeated until the assignment of the total flow rate: 130 is completed. The basic operations are (1) to (10) as described above, but in practice it is necessary to operate the reverse flow rate.

The "shortest problem" includes a "Dijkstra method", the "maximum flow problem" includes a "labeling method", and the "least cost flow problem" includes a solution such as a "primal dual method". This time, you can use these methods. The mathematical expression of "network programming" is generally expressed as the following equation.

(Objective function)

(Equation 7)

(Constraints)

(Equation 8)

Here, total cost: Z arc flow rate: Xk Upper and lower limits of capacity of “arc: k”: Uk, Vk Cost per unit flow rate of “arc: k”: Ck Source node: s Sink node: t Total flow from “node: s” to “node: t”: q M: all arcs

A method for applying such a network planning method to a pulp and paper manufacturing facility will now be described based on embodiments. FIG. 2 shows an example of a pulp and paper production facility.
This is a manufacturing process shown as an example in the section of the prior art. In FIG. 2, reference numeral 12 denotes a raw material (pulp) production facility, which indicates that there are five units. Reference numeral 13 denotes a cushion chest, that is, a raw material storage facility, in which five are installed. A blend chest 14 mixes raw materials supplied from one or more cushion chests 13. The mixed raw material is sent to five types of product processes 15 and the product (10 in the figure)
Seed) 16 is completed.

FIG. 3 shows a network flow for scheduling a production plan of such a manufacturing process.
Because this network flow has many nodes and is complicated,
FIG. 4 shows a partially simplified model for explanation.

In FIG. 4, reference numeral 20 denotes a node for passing the total inflow power. Reference numeral 21 denotes a node group for the power type, which is a node group on the network including three types of nodes for selecting the power type, that is, three types of nodes: thermal power, hydropower, and purchased power.
Reference numeral 22 denotes a node group having nodes for the raw material production facilities. FIG. 4 shows that there are five nodes (see the raw material production facilities N1 to N5 in FIG. 2). Reference numeral 23 denotes a node group having nodes related to the cushion chest, and the example of FIG. 4 assumes that there is no delay in the production time of the raw material (pulp). Reference numeral 24 denotes a node for passing the total power flowing out. Reference numeral 25 denotes a node for passing an amount of electric power used in a facility other than the raw material manufacturing facility, but the arc and the node are omitted in FIG.

In such a network, the node groups 20 to 25 are connected by an arc according to the relationship of the production process. Node group 2
The nodes in 1 are independent of each other and are not connected to each other by an arc.

The process of the production facility shown in FIG. 2 is represented by a network as shown in FIG. FIG.
In FIG. 5, A0-1, A0-2 to
Table 1 shows the capacity constraint and unit cost condition of each arc represented by each group of A11. Here, it should be noted that the raw material storage amount and the raw material amount to be supplied are converted into the corresponding power consumption in advance using the power consumption unit of raw material (pulp) production: EUn. is there. By performing this processing, the optimal production schedule can be solved as the minimum cost flow problem of the network planning method.
Manufacturing-related parameters described in the conventional example (PPg, m, t: required amount of paper product production, Hg, n, m: raw material (pulp) blending ratio of paper product)
Is used, the total flow rate of the minimum cost flow problem of the network model of FIG.
It can be represented by the following equation (1). If this network model is solved as a minimum cost flow problem, the flow rate flowing through each arc per unit time is determined, and the optimal production schedule can be determined.

[0050]

[Table 1]

Here, P _{n, Delay} : the amount of each raw material (pulp) produced with a time delay in the beginning of the new production schedule P _{n, Next} : in the beginning of the next production schedule Each raw material (pulp) to be produced in advance with a time delay
Planned amount C _{n, Start} : Initial raw material storage amount of each raw material storage tank (cushion chest) at the start of production schedule C _{n, Final} : Planned final raw material storage amount of each raw material storage tank (cushion chest) at the end of production schedule E _p: the amount of the P-band power purchased E _op: the amount of the OP band purchased electricity E _N: usage of the N band purchased power E _B: the amount of private power generation thermal power E _H: private power generation hydroelectric power usage other Use the symbols described in the prior art.

FIG. 6 is a Gantt chart summarizing product production conditions for obtaining an optimal production schedule. The production condition at the time sandwiched between the two in FIG. 6 corresponds to the state of the network sandwiched between the two in FIG. Based on the production conditions shown in FIG. 6, the type of the product and the production amount are determined in advance as input data, and given to the optimal production schedule creation device.

FIGS. 7 to 9 show the optimum solution (FIG. 7 is the production amount of each raw material, FIG. 8 is the storage amount of the cushion chest, and FIG. 9 is the electric power consumption) obtained by the method of the present invention. This is represented by a graph on the horizontal axis.

Next, an optimum production schedule creating apparatus to which the present invention is applied will be described with reference to FIG. FIG. 10 shows an example of the system configuration of the optimum production schedule creation device. A general-purpose computer such as a personal computer or a workstation is used as the optimal production schedule creation device. In FIG. 10, the CPU 1 controls the operation of the entire system, and is involved in the creation of a schedule, and acquires the value of a manufacturing element (parameter) that minimizes the total power amount of the production equipment.

The system memory 2 includes a ROM that stores a program that defines system processing to be executed by the CPU 1, data used for system processing, and a CP.
It is composed of a RAM and the like for temporarily storing data used for the operation of U1. Hard Disk Storage (HD) 3
Stores a system program other than the system program stored in the ROM and a program for creating a production schedule according to the present invention. Also, information for visually displaying condition values, fixed set values, created production schedules, and production schedules in the flowchart of FIG. 2 is stored.

The input interface 4 is a pointing device (hereinafter, a mouse) such as a keyboard 5 and a mouse.
6 is transferred to the CPU 1. Keyboard 5
Also, data used for creating a production schedule can be input from the mouse 6.

The output interface 7 is the system memory 2
Display information and print information stored in the display device (hereinafter, C)
RT) 8 and the printer 9. The CRT 8 and the printer 9 visually display the schedule creation result and the transition of the power consumption in time series. Communication interface 10 is LAN
11 to receive data sent from another system or to send the created production schedule to another system.

FIG. 11 shows a process related to creating a production schedule executed in such a system configuration.
Note that this processing procedure is an example, and the processing procedure can be any desired one according to the usage pattern of the user.

When receiving an operation instruction from the keyboard 5, the CPU 1 identifies the content of the operation instruction and performs a process according to the instruction. The condition values and set values of various parameters of the production equipment shown in FIG. 2 and Table 1 are prepared in advance,
The data stored in the HD 3 is updated by inputting data relating to the production time and production amount of the product and the above-mentioned parameters (step S1 → S2 → S1).
0). The number of raw material production facilities, the number of cushion chests, and the like are generally preferably set to fixed values, but can be variably set on a network model. For this reason, it is preferable to construct a program so that not only the value of the condition of the network model but also the value of the number of facilities can be changed.

When the operation instruction input from the keyboard 5 is an instruction to create a schedule, the CPU 1 determines a path (node path) that minimizes the power cost of the network model using a known solution, Also, the values of the respective parameters at that time are stored in the HD 3 (steps S 1 → S 2 → S 3 → S 31). In addition, if there is a need for modification, the user modifies the automatically created schedule interactively using the keyboard 5 and mouse 6 (step S32).

After that, the output form of the creation schedule, for example, print output, display output, etc., the format form,
For example, receiving an instruction in a form as shown in FIGS.

Here, the processing time of the computer when the same problem is obtained by the optimization method using the commonly used “linear programming (LP)” and when it is calculated by the “network programming” Are compared in Table 2.

The "network planning method" can perform calculations two to three times faster in the calculation time, and requires only 1/3 to 1/7 times the memory used by the computer. It can be seen that calculations can be performed within practical time with a small computer. Also, the "network planning method" is known as a method that exerts an effect in terms of processing speed as the calculation scale of the target increases, and the calculation time for the problem scale is "LP". While the number increases roughly in proportion to (the number of constraint equations) ³ × (the number of unknowns), the “network planning method”
It has been found to be less than that.

[0064]

[Table 2]

It can be seen that the use of the "network planning method" for such a large-scale optimal problem enables a high-speed calculation and a calculation execution within a practical time.
The following example can be implemented in addition to this embodiment.

1) In many cases, the paper product production amount based on the prepared raw material production schedule does not match the originally planned paper product production amount. Therefore, an error between the production results obtained so far and the production plan amount of the schedule can be calculated, and the error can be considered as basic data for creating a new schedule.

2) Network Model When displayed in the form of FIG. 3 or FIG. 4, the contents of the production equipment can be known. Further, a value currently set for the network model may be displayed by an instruction from the keyboard 5.

[0068]

As described above, according to the first to fifth aspects of the present invention, it is possible to quickly create a production schedule using a solution of a network planning method by modeling a production facility in a network form. Becomes Also,
It is possible to create an optimal production schedule that takes into account changes in parameters such as the unit price of electric power over time in association with the production schedule of (paper) products.

In particular, according to the first aspect of the present invention, by converting a parameter into a numerical value of the same kind of parameter,
It can be modeled as a minimum cost flow problem in network planning, and processing time can be reduced.

According to the second, third, and fourth aspects of the present invention, a paper pulp manufacturing facility in which the power cost accounts for a relatively large proportion of the manufacturing cost, there are several types of power supply means, and the power unit price changes with time. It is possible to provide an optimum production schedule automatic creation method and apparatus suitable for the production schedule.

It should be noted that, depending on the ability of the person in charge of planning, the optimized operation schedule is compared with a schedule created by a human system, and when the embodiment described in the text is actually applied, Simulations have also shown that a reduction of about 3% in total plant operating power costs can be expected. Since the optimal schedule can be automatically created, the operating power cost can always be stably reduced without being affected by the ability and physical condition of the person in charge. Further, according to the invention of claim 5, it is possible to cope with a change in manufacturing equipment.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining a network and a minimum cost flow problem.

FIG. 2 is a flowchart illustrating a configuration of a production facility.

FIG. 3 is an explanatory diagram showing a network model corresponding to a production facility.

FIG. 4 is an explanatory diagram showing a part of a network model.

FIG. 5 is an explanatory diagram showing positions of arcs having the restrictions shown in Table 1 on a network.

FIG. 6 is an explanatory diagram showing a part of contents of production conditions (preconditions) of a product for obtaining an optimal production schedule.

FIG. 7 is an explanatory diagram showing a part (raw material production amount) of a calculation result of an optimal production schedule.

FIG. 8 is an explanatory diagram showing a part (storage amount of a storage tank) of a calculation result of an optimal production schedule.

FIG. 9 is an explanatory diagram showing a part (power consumption) of a calculation result of an optimal production schedule.

FIG. 10 is a block diagram showing a system configuration of an optimum production schedule creation device.

FIG. 11 is a flowchart illustrating a processing procedure of the optimum production schedule creating device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 CPU 2 System memory 3 HD4 Input interface 5 Keyboard 6 Mouse 7 Output interface 8 CRT 9 Printer 10 Communication interface 11 LAM 12 Raw material (pulp) production equipment 13 Raw material storage equipment (cushion chest) 14 Blend chest 15 Product process 16 Product 20 ~ 25 node groups

Claims (5)

[Claims] 1. A production schedule is created by quantifying a plurality of manufacturing elements that affect manufacturing as parameters for each type, and determining, by a computer, parameter values that optimize an evaluation criterion determined by a combination of the parameters. In the method for automatically creating an optimal production schedule, a network model in which interconnected manufacturing elements among the plurality of manufacturing elements are connected in a network form is assumed, and a path on the network model where the evaluation criterion is optimal. In order to obtain the value of the parameter at that time, the solution of the minimum cost flow problem of the network programming is used, and in order to solve the minimum cost flow problem, the values of the parameters corresponding to the plurality of manufacturing elements are the same. Converted to a kind of parameter numerical value, the optimal path and the parameter at that time Acquires every unit time, based on the value of the acquired path and parameters, automatically creating the optimal production schedule, characterized in that to create the production schedule. 2. The automatic production schedule creation method according to claim 1, wherein said evaluation criterion has a minimum cost. 3. The method according to claim 1, wherein the production is paper pulp.
In the production of the paper pulp, a plurality of types of raw materials (pulp) are prepared, and the raw materials are blended to produce a plurality of types of products (paper).
The raw material can be stored in a storage tank for each type, and in a manufacturing process in which a plurality of types of power having different power unit prices are selectively supplied, a production time and a production amount of a product to be produced are input. Given as data, the evaluation criterion is the power minimum cost, the same type of parameter is calculated as power consumption, raw material production per unit time, power consumption and storage capacity of the storage tank are obtained, and the optimal production schedule Automatic creation method of the optimal production schedule characterized by creating. 4. A production schedule is created by quantifying a plurality of production factors affecting production as parameters for each type, and determining, by a computer, parameter values that optimize an evaluation criterion determined by a combination of the parameters. In the optimal production schedule creation device, a network model in which interconnected manufacturing elements among the plurality of manufacturing elements are connected in a network form is determined in advance, and a path on the network model where the evaluation criterion is optimal. In order to obtain the value of the parameter at that time, the solution of the minimum cost flow problem of the network programming is used, and in order to solve the minimum cost flow problem, the values of the parameters corresponding to the plurality of manufacturing elements are the same. Converted to a kind of parameter numerical value, the optimal path and the parameter at that time Optimum production, comprising: first information processing means for acquiring a value for each unit time; and second information processing means for creating a production schedule based on the acquired path and parameter values. Schedule creation device. 5. The optimum production schedule creation device according to claim 4, wherein the structure of the network model can be variably set.