JP4328071B2 - Optimization simulation apparatus, optimization simulation method, and program - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention provides, for example, an optimization simulation apparatus, an optimization simulation method, and an optimization simulation method that are used in an apparatus or a system for solution search that can optimally pick up an order coil (ordered product) from a parent coil (base material) in the steel industry. It is about the program.
[0002]
[Prior art]
Conventionally, for example, various optimization problem solving methods have been used in operations such as performing complex scheduling in a production plan of the manufacturing industry, or arranging elements within a certain range.
[0003]
For example, in the steel industry, a solution search is performed under a plurality of constraints for an optimal solution for order manufacturing, such as what, in what order, and how much to produce for a certain order. .
[0004]
As an example of an optimization problem in the steel industry, there is a plate cutting optimization problem for optimally cutting a coil according to an order (hereinafter referred to as “shipping coil” or “order coil”) from a parent coil.
[0005]
Specifically, first, when producing an order coil, a certain parent coil is divided into a predetermined width and length and cut according to the order. The constraint conditions at this time include the outer diameter size of the order coil itself, the weight, the weight constraint coming from equipment such as a crane, the order appropriation weight, and the limit on the plate size received from the size of the parent coil.
[0006]
Here, coils having various widths and lengths are prepared in advance as the parent coil. Therefore, in order to optimally produce a custom coil from a parent coil, for example, an appropriate parent that satisfies the above-mentioned constraints, for example, requires less cutting equipment replacement, parent coil replacement, and less waste area. The coil must be selected from a plurality of parent coils. That is, it is necessary to search for an optimal solution.
[0007]
[Problems to be solved by the invention]
However, since the solution search for solving the optimization problem as described above has been conventionally performed by human experience, there are the following problems.
[0008]
For example, in the steel industry, as shown in FIG. 6, the administrator 901 gives a slit instruction to the worker 902 using the instruction sheet 903. Thereby, the worker 902 performs the slit work in accordance with the slit instruction.
[0009]
Here, the slitting operation refers to an operation of cutting out order coils (shipping coils) a1 to a3, b1 to b3, and c1 to c3 from the parent coil A.
Therefore, the administrator 901 has information on the parent coil A (information on the width of the parent coil A of 900 mm) and information on the order coils a1 to a3, b1 to b3, and c1 to c3 (width 700mm of the order coils b1 to b3, and The order coil b1 to b3 having a width of 700 mm is assigned to the parent coil A having a width of 900 mm by the information on the width 100 mm of the order coils a1 to a3 and c1 to c3). A slit instruction (903) for assigning the order coils a1 to a3 and c1 to c3 having a width of 100 mm and dividing them into three is determined.
[0010]
At this time, the administrator 901 needs to determine the best possible slit instruction (903).
FIGS. 7A to 7C show other examples of the slit instruction (903).
[0011]
In the slit instruction shown in FIG. 7A, the order coils b1 to b3 are assigned to the parent coil B having the same width (700 mm) as the order coils b1 to b3 and divided into three, and the order coils a1 to a3 and c1 are divided. The order coils a1 to a3 and c1 to c3 are assigned to the parent coil C having substantially the same width as the total width (600 mm) of .about.c3. In this slit instruction, the order coils b1 to b3 are ordered from the parent coil B. In order to plate the order coils a1 to a3 and c1 to c3 from the parent coil C having different width sizes, it is necessary to perform blade replacement and coil replacement. Therefore, production efficiency falls.
[0012]
The slit instruction shown in FIG. 7 (b) is the same instruction as the pattern of planing as shown in FIG. 6, but the width of the parent coil is the order coils a1-a3, b1-b3, c1- Since it is wider than the total width of c3 (900 mm), waste is generated correspondingly, and the yield decreases.
[0013]
The slit instruction shown in FIG. 7 (c) is also the same instruction as the planing pattern as shown in FIG. 6 above, but since the division into three is made into four, the order of the same weight However, since the number of divisions increases, that is, many small coils are shipped, work efficiency after packing is reduced accordingly.
[0014]
Therefore, a good slit instruction is not an instruction as shown in FIGS. 7A to 7C, but a small number of blade replacement / coil replacement (number of sets) as shown in FIG. It can be said that the instruction satisfies the condition that the scrap area is small and the shipping coil is large.
[0015]
As a constraint for satisfying the above condition, for example, as shown in FIG.
1. The size of the order coil (shipment coil) must be the size specified by the customer.
2. Since it will be shipped with multiple coils, it will exceed the weight actually ordered, but this overload is within the allowable range.
3. As a matter of course, the size of the parent coil is larger than the order coil (shipment coil).
There are three restrictions.
[0016]
Therefore, for example, as shown in FIG. 9, the administrator 901 takes into account the constraints shown in FIG.
1. A wide custom coil (main order, here 700 mm coil) is used for the parent coil.
2. If there is another small order coil (in this case, a 100 mm coil) that fits the remaining width of the parent coil, this is assigned to the parent coil. At this time, when the width of another order coil does not correspond to the remaining width of the parent coil, the order coil is assigned to the parent coil having another width.
Do the work.
[0017]
Specifically, for example, as shown in FIG. 10, there are three parent coils A, B, and C, and custom coils a to f having different widths and lengths are produced from these parent coils A, B, and C. First, the manager 901 allocates three order coils a to c to the parent coil A. As a result, the order coils a and b can be removed from the parent coil A, but the order coil c cannot be removed from the parent coil A, that is, the width and length of the order coil c are not suitable for the parent coil A. 901 assigns the order coil c to the next parent coil B.
[0018]
Next, the manager 901 further fills the order coils d and e with respect to the parent coil B filled with the order coil c. As a result, the order coils c and d can be removed from the parent coil B, but the order coil e cannot be removed from the parent coil B, that is, the width and length of the order coil e are not suitable for the parent coil B. 901 assigns the order coil e to the next parent coil C.
[0019]
Then, the administrator 901 fills the last order coil f with respect to the parent coil B filled with the order coil e.
[0020]
Therefore, as a slit instruction (optimum solution for plate cutting), order coils a and b are assigned to the parent coil A, order coils c and d are assigned to the parent coil B, and An instruction to assign the order coils e and f will be obtained.
[0021]
However, as a matter of course, as described above, when a person has made an optimum slit instruction, that is, a decision to obtain an optimum solution, especially when the number of orders increases, it will not be in time for the order or waste. As a result, there arises a problem such as the generation of a remaining coil that must be disposed of, and it is practically impossible to derive an optimal solution.
[0022]
In particular, the size of the order coil is restricted by the weight limit of the crane on the customer side, one hoop length (coil length at the time of shipment) is restricted, and the hoop length is allowed within an arbitrary range. Therefore, it is naturally determined from which fixed coil the custom coil can be picked up, but from the various parent coils with different widths and lengths, an appropriate parent coil is searched without yield. Is very difficult.
[0023]
Therefore, a configuration has been proposed in which a solution search is performed by a computer system equipped with an optimization engine.
[0024]
However, even in the configuration using the above-described computer system, the solution search model and the constraint conditions are basically the same. As a result, the solution space is tried manually as described above. There is no change to the method of searching for the optimal solution through mistakes, and even if a better solution can be found because the calculation is faster than humans, it is not possible to derive the optimal solution from a solution space close to infinity. However, it cannot always be done at high speed.
[0025]
Therefore, the present invention was made to eliminate the above-mentioned drawbacks, and is an information processing apparatus, an information processing system, and an optimization calculation support capable of deriving an optimal solution at high speed under complicated constraints. It is an object to provide a method, a computer-readable storage medium storing a program for implementing the method, and the program.
[0026]
[Means for Solving the Problems]
Under such an object, the optimization simulation apparatus according to the present invention adds the length information of a plurality of base materials inputted through the information input means for inputting information, and has a virtual length having one virtual length. Generating base material information and storing the base material aggregation means in the memory; information on the virtual base material stored in the memory; information on a plurality of orders input via the information input means; And an optimization processing means for deriving an optimal solution for allocation from a combination of a plurality of orders allocated to the virtual base material.
[0027]
Further, the optimization simulation method of the present invention is such that the optimization simulation apparatus adds up the length information of a plurality of base materials inputted via the information input means for inputting information to obtain one virtual length. A base material aggregation step for generating virtual base material information to be stored in a memory; and the optimization simulation apparatus inputting the virtual base material information stored in the memory via the information input means And an optimization processing step of deriving an optimal solution for assignment from a combination of assignment of the plurality of orders to the virtual base material based on the information on the plurality of orders.
[0028]
Further, the program of the present invention is a computer program that adds information on the lengths of a plurality of base materials inputted via an information input means for inputting information to the virtual base material having one virtual length. Based on base material aggregation means for generating and storing in a memory, information on the virtual base material stored in the memory, and information on a plurality of orders input via the information input means, It is made to function as an optimization processing means for deriving an optimal solution for allocation from a combination in which a plurality of orders are allocated to the virtual base material.
[0031]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0032]
The present invention is applied to, for example, an optimization calculation support apparatus 100 as shown in FIG.
The optimization calculation support device 100 according to the present embodiment is a device having a function of obtaining an optimum solution of plate cutting for producing a plurality of order coils from a plurality of parent coils (base materials) in the steel industry, for example. In particular, it is configured to efficiently obtain an optimal solution by narrowing the solution space with a virtual coil obtained by integrating a plurality of parent coils.
Hereinafter, the configuration and operation of the optimization calculation support device 100 according to the present embodiment will be specifically described.
[0033]
<Configuration of Optimization Operation Support Device 100>
As shown in FIG. 1, the optimization calculation support apparatus 100 includes an information input unit 110 that inputs information on a parent coil, an order coil, and the like, and an optimized solution search process that obtains an optimal solution based on input information from the information input unit 110. Unit 120, a solution output unit 130 for outputting the optimum solution obtained by the optimized solution search processing unit 120, and an actual work instruction (slit instruction) based on the optimum solution output by the solution output unit 130 The post-processing unit 140 is included.
[0034]
The optimized solution search processing unit 120 aggregates a plurality of parent coils to form a virtual coil, and optimizes to obtain an optimal solution by assigning orders to the virtual coils obtained by the parent coil aggregating unit 121 The simulation engine 122 is included.
[0035]
Note that any engine can be applied as the optimization simulation engine 122. In the present embodiment, as an example, by using optimization package software such as “ilog solver” of ilog France, all combinations within the constraints are executed, and an optimal solution is derived. .
[0036]
The components 110 to 140 shown in FIG. 1 are implemented by a computer function 200 as shown in FIG.
[0037]
As shown in FIG. 2, the computer function 200 includes a CPU 201, a ROM 202, a RAM 203, a keyboard controller (KBC) 205 of a keyboard (KB) 209, and a CRT controller (CRT controller (CRT) 210 as a display unit). CRTC) 206, hard disk (HD) 211 and disk controller (DKC) 207 of flexible disk (FD) 212, and network interface controller (NIC) 208 for connection to network 220 via system bus 204 It is the structure connected so that communication was possible mutually.
[0038]
The CPU 201 comprehensively controls each component connected to the system bus 204 by executing software stored in the ROM 202 or the HD 211 or software supplied from the FD 212.
That is, the CPU 201 reads out a processing program (a program including the optimization simulation software described above) from the ROM 202, the HD 211, or the FD 212 and executes the processing program according to a predetermined processing sequence. Control to realize
[0039]
The RAM 203 functions as a main memory or work area for the CPU 201.
The KBC 205 controls an instruction input from the KB 209 or a pointing device (not shown).
The CRTC 206 controls the display of the CRT 210.
The DKC 207 controls access to the HD 211 and the FD 212 that store a boot program, various applications, an edit file, a user file, a network management program, a predetermined processing program in the present embodiment, and the like.
The NIC 208 bidirectionally exchanges data with devices or systems on the network 240 as necessary.
[0040]
<Operation of Optimization Computing Support Device 100>
In this embodiment, it is assumed that there is a welding machine for the sake of simplicity. Therefore, it is not always necessary to plate the order coil from the parent coil with the length specified in the order. For example, it is possible to plate a plurality of parent coils according to the order across the parent coils, and then weld these to the final coil. It can be shipped as a custom coil.
[0041]
In the optimization calculation support apparatus 100 shown in FIG. 1, first, the information input unit 110 receives information about the parent coil (parent coil information), information about the order (order information), constraint condition information, priority condition information, and the like. input.
[0042]
The parent coil information is information relating to the width and length of the parent coil.
The order information is information related to the width and length of the order coil.
The constraint condition information is an indispensable condition, such as the shipping coil (order coil) length Min-Max constraint, the order charge equivalent Min-Max constraint, and the set width / length constraint as described with reference to FIG. It is information about.
The priority condition information is information relating to a condition that can be changed and that determines whether the result is good or bad. Examples of the priority condition include conditions such as a smaller number of sets, a smaller waste area, and a larger shipping coil.
[0043]
The optimization solution search processing unit 120 obtains an optimal solution based on the input information from the information input unit 110 by the optimization simulation engine 122. At this time, in this embodiment, in particular, the parent coil aggregation unit 121 performs a plurality of operations. The solution space is narrowed by consolidating the parent coil (actual product) having the same width or a similar width and assuming a virtual coil.
[0044]
Specifically, for example, as illustrated in FIG. 3, the parent coil aggregation unit 121 includes, as an example, a plurality of parent coil information having various widths and lengths input via the information input unit 110. Based on the same parent coils A, B, and C , the lengths of these parent coils are summed to generate one virtual coil α. This is because in the actual slitting operation (see FIG. 6 above), since the plate cutting operation is performed while welding a plurality of parent coils, the actual parent coils A, B, C (actual product) are not welded. This is because it is necessary to assign an order coil. This welded result corresponds to the virtual coil α.
[0045]
In this way, assuming the finite length of the parent coils A, B, C as one infinite length coil (virtual coil α) with the coil width type (per width), the hoop length of the custom coil (at the time of shipment) An appropriate parent coil can be selected by paying attention only to the width of the order coil without being limited by the coil length.
[0046]
Therefore, as shown in FIG. 4, the virtual coil α obtained from the parent coils A, B,... And the virtual coil β obtained from the parent coils E, F,. Information of these virtual coils α and β is input to the optimization simulation engine 122.
[0047]
As shown in FIG. 4, the optimization simulation engine 122 includes virtual coil information from the parent coil aggregating unit 121 and information input by the information input unit 110 (information on order coils a, b, c, constraint condition information, and By using the priority condition information and the like, a predetermined algorithm is executed to obtain an optimal solution for the virtual coils α and β and the order coils a, b, and c.
[0048]
Since the optimization simulation engine 122 obtains a solution in which an order coil is assigned to a virtual coil in which the parent coils are aggregated instead of the parent coil that is the actual product as in the past, the solution space is narrowed. In this state, the solution search is performed.
[0049]
That is, in the present embodiment, due to the aggregation of the parent coil, for a certain kind of coil width,
Virtual coil: Custom coil = 1: N
Since the solution space is narrowed, the optimal solution can be obtained efficiently.
[0050]
In FIG. 3, the width of the coil C having the narrowest width among the parent coils A, B, and C is the width of the virtual coil α (temporary width). However, the present invention is not limited to this. Depending on the situation, the optimum width may be set as the width of the virtual coil α.
Specifically, first, the widths of the parent coils A, B, and C are basically the same or close to each other. However, as shown in FIG. , B, and C, the width of the virtual coil α obtained from these parent coils A, B, and C may be any coil width among the parent coils A, B, and C. This is a combined solution of assigning a plurality of order coils to a certain virtual coil. In practice, for example, the post-processing unit 140 then performs the virtual solution. This is to execute a specific process of assigning which order coil to which parent coil in the parent coil (actual product) constituting the coil.
[0051]
The solution obtained by the optimized solution search processing unit 120 as described above is displayed and output by the solution output unit 130 to, for example, a user (such as an administrator).
[0052]
For example, the post-processing unit 140 executes processing for issuing a slit instruction as described with reference to FIG. 6 based on the solution output by the solution output unit 130.
[0053]
Specifically, for example, as shown in FIG. 4 (see the dotted line “solution”), when a solution that assigns the order coil a to the virtual coil α is obtained as one solution, the post-processing unit 140 determines (selects) from which parent coil the order coil a is to be cut based on information such as width and length for the actual parent coils A, B,... Constituting the virtual coil α. To do.
[0054]
The determination in the post-processing unit 140 is issued to the worker as a slit instruction. As a result, based on the slit instruction, the operator can plate the corresponding custom coil from the corresponding parent coil, weld it if necessary, and finally produce a coil of the size indicated by the customer. Ship.
[0055]
<Effects of this embodiment>
Conventionally, as described with reference to FIG. 10 above, when a good solution result is not obtained in a manual solution search, a different solution is searched by backtracking. In this form, the solution search is performed for virtual coils that are aggregated based on the coil width of the parent coil, so the constraints are greatly relaxed, and the solution search can be performed at high speed without backtracking across multiple parent coils. Yes.
[0056]
In the case where the length of the order coil is long, conventionally, it is necessary to divide the order coil and search for a parent coil that can be picked up for each of the divided order coils. Since the virtual coil has an infinite length, the entire order coil is always included in the virtual coil. Therefore, the solution space can be narrowed accordingly, and the solution search can be performed at a very high speed.
[0057]
For example, as shown in FIG. 5, as a result of the first solution search, all of one order coil a cannot be assigned to one virtual coil α, and a remaining (a2) appears in order coil a. In this case, the solution search may be performed again for the remaining coil a2.
[0058]
In the present embodiment, it is assumed that there is a welder. However, when there is no welder, the hoop length of the custom coil (the coil length at the time of shipment) is fixed, and this is used as a virtual coil. You may make it perform the solution search to allocate. in this case,
Virtual coil: Custom coil = N: N
This is effective because the order division can be reduced by using the virtual coil.
[0059]
Also, an object of the present invention is to supply a storage medium storing software program codes for realizing the functions of the host and terminal of the present embodiment to a system or apparatus, and the computer of the system or apparatus (or CPU or MPU). Needless to say, this can also be achieved by reading and executing the program code stored in the storage medium.
In this case, the program code itself read from the storage medium realizes the functions of the present embodiment, and the storage medium storing the program code and the program code constitute the present invention.
As a storage medium for supplying the program code, ROM, flexible disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD-R, magnetic tape, nonvolatile memory card, and the like can be used.
Further, by executing the program code read by the computer, not only the functions of the present embodiment are realized, but also an OS or the like running on the computer based on an instruction of the program code performs actual processing. It goes without saying that a case where the function of this embodiment is realized by performing part or all of the above and the processing thereof is included.
Further, after the program code read from the storage medium is written to the memory provided in the extension function board inserted in the computer or the function extension unit connected to the computer, the function extension is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the board or function expansion unit performs part or all of the actual processing, and the functions of the present embodiment are realized by the processing.
[0060]
【The invention's effect】
As described above, in the present invention, in order to optimally assign a plurality of order products (coiled order metal, etc.) to a plurality of base materials (coiled metal base material, etc.) The materials are aggregated to generate a virtual base material, and a plurality of ordered products are assigned to the virtual base material.
For example, N base materials having substantially the same width are connected to generate M (M <N) virtual base materials, and a plurality of ordered products are assigned to the M virtual base materials. Thereby, a plurality of ordered products can be assigned to an infinite length virtual base material.
[0061]
With the above configuration, the solution space can be narrowed even under complicated constraints, and an appropriate solution search can be performed efficiently.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an optimization calculation support apparatus to which the present invention is applied.
FIG. 2 is a block diagram illustrating a configuration of a computer function included in the optimization calculation support apparatus.
FIG. 3 is a diagram for explaining an optimization solution search processing unit of the optimization calculation support apparatus.
FIG. 4 is a diagram for explaining virtual coils obtained by the optimized solution search processing unit.
FIG. 5 is a diagram for explaining a case where a balance is left in the order coil as a result of the assignment of the order coil to the virtual coil in the optimization calculation support apparatus.
FIG. 6 is a diagram for explaining slit instructions and work in the steel industry.
FIG. 7 is a diagram for explaining an example of the slit instruction.
FIG. 8 is a diagram for explaining a restriction at the time of instructing the slit.
FIG. 9 is a diagram for explaining assignment work for a parent coil of a conventional order coil.
FIG. 10 is a diagram for describing a specific example of assignment work for a parent coil of the order coil.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 100 Optimization calculation support apparatus 110 Information input part 120 Optimization solution search process part 121 Parent coil aggregation part 122 Optimization simulation engine 130 Solution output part 140 Post-processing part

Claims (7)

Aggregate the base materials that generate the information of the virtual base material having one virtual length by summing the length information of multiple base materials input via the information input means for inputting information, and store it in the memory Means,
A combination in which a plurality of orders are assigned to the virtual base material based on the information on the virtual base material stored in the memory and information on a plurality of orders input via the information input means An optimization simulation device comprising an optimization processing means for deriving an optimal solution for assignment from the system.   The optimization processing means includes information on the virtual base material stored in the memory, information on a plurality of orders input via the information input means, and constraints input via the information input means. The optimization simulation apparatus according to claim 1, wherein an optimal solution for assignment is derived from a combination in which a plurality of orders are assigned to the virtual base material based on a condition.   The metal base material is included in the plurality of base materials, and length information of the metal base material is included in the length information of the plurality of base materials. Optimization simulation equipment. The optimization simulation device generates the information of a virtual base material having one virtual length by summing up the length information of a plurality of base materials input via the information input means for inputting information, and the memory A base material aggregation step to be stored in
Based on the information on the virtual base material stored in the memory and information on a plurality of orders input via the information input means, the optimization simulation device applies the virtual base material to the virtual base material. An optimization process step for deriving an optimal solution for allocation from a combination of multiple orders;
An optimization simulation method characterized by comprising:   In the optimization processing step, the optimization simulation apparatus includes information on the virtual base material stored in the memory, information on a plurality of orders input via the information input unit, and the information input unit. 5. An allocation optimal solution is derived from a combination in which a plurality of orders are allocated to the virtual base material based on information and constraint conditions related to the plurality of orders input via the network. Optimization simulation method described in 1. Computer
Aggregate the base materials that generate the information of the virtual base material having one virtual length by summing the length information of multiple base materials input via the information input means for inputting information, and store it in the memory Means,
A combination in which a plurality of orders are assigned to the virtual base material based on the information on the virtual base material stored in the memory and information on a plurality of orders input via the information input means A program characterized by functioning as an optimization processing means for deriving an optimal solution of assignment from the program.   The optimization processing means includes information on the virtual base material stored in the memory, information on a plurality of orders input via the information input means, and constraints input via the information input means. The program according to claim 6, wherein an optimal solution for assignment is derived from a combination of a plurality of orders assigned to the virtual base material based on a condition.

Images