JP2007183817A - Scheduling device, scheduling method, scheduling program, and recording medium with the program recorded thereon - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device which performs highly precise production scheduling. <P>SOLUTION: A scheduling device, which schedules the production of an object to be produced through a plurality of processes, is provided with: a storage means which stores process connection information for setting relations of process connection-order, block flow information for setting a moving route of each block included in the processes, work-term information for setting a work term of each process in each block, and restrictive conditions of each process; an interpretation means which sorts the processes in ascending order from the downstream to the upper stream based on the information stored in the storage means; a model-preparing means which makes a scheduling model based on process data after sort which are obtained by the interpretation means; a schedule plan-preparing means which optimizes schedules for each scheduling model obtained by the model-preparing means; and an outputting means which outputs the result of the scheduling obtained by the schedule plan-preparing means. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a scheduling device, a scheduling method, a scheduling program, and a recording medium on which the program is recorded, and more particularly to a scheduling device, a scheduling method, a scheduling program, and the program for realizing highly accurate production scheduling. The present invention relates to a recording medium.

  Conventionally, there is a system for optimizing and scheduling a schedule plan in a work place or the like when producing a custom-made product at a shipyard, a factory, or the like (see, for example, Patent Document 1).

In addition, in order to perform scheduling in consideration of various constraints in the factory, based on the constraints in the factory, by running a simulation model that configures the virtual factory, perform a simulation to manufacture things, There is a method of outputting a schedule based on the simulation result (see, for example, Patent Document 2).
JP 2003-131721 A JP 2002-163012 A

  However, the technique disclosed in Patent Document 1 schedules custom-made products, but does not consider the schedule for each process. In addition, the technique disclosed in Patent Document 2 is to perform scheduling by actively using simulation, but performs scheduling in a batch, such as a work place (process) group. It is not based on a subset. Therefore, high-precision scheduling has not been achieved for partial processes that are particularly important, such as bottlenecks in the entire factory and complicated constraints.

  In addition, there is a method for obtaining an exact solution using a branch and bound method, etc., but this method is intended for, for example, large-scale one-order products, and is not practical for large-scale production scheduling. There wasn't.

  The present invention has been made in view of the above problems, and provides a scheduling device, a scheduling method, a scheduling program, and a recording medium on which the program is recorded, for performing highly accurate production scheduling. Objective.

  In order to achieve the above-described object, the present invention includes a process connection information for setting a connection order relationship of the processes and a process connection information for scheduling the production of a production object including a plurality of processes. Block flow information for setting the movement path of each block, work schedule information for setting the work schedule in each process of each block, and storage means for storing the constraint conditions for each process, and the storage means. Interpretation means for rearranging the processes from the downstream information in the order of going back to upstream, model creation means for creating a scheduling model based on the rearranged process data obtained by the interpretation means, and obtained by the model creation means Schedule planning means for optimizing the schedule for each scheduling model, and the schedule planning means And an outputting means for outputting a scheduling result to be. Thereby, highly accurate production scheduling can be realized. In addition, high-precision large-scale scheduling in a practical time can be realized.

  Furthermore, based on the scheduling result obtained by the schedule planning means, there is provided a simulation means for performing a simulation for every process or a preset number of processes and outputting the result to the storage means or the output means. Is preferred. Thereby, the scheduling content can be confirmed along with the change of time, and it is possible to make it easy for the user to grasp the accuracy. Accordingly, the user or the like can accurately grasp the process of the bottleneck portion expected in the future work process, and can perform quick and reliable correction (work change) from the contents.

  Furthermore, it is preferable to have plan data correction means for correcting the scheduling result obtained by the schedule planning means based on the simulation result obtained by the simulation means. As a result, it is possible to obtain scheduling results reflecting the know-how and the like possessed by experts who cannot be described as input conditions.

  Furthermore, it is preferable that the model creation unit executes scheduling by rounding a work period of each work obtained by the interpretation unit. Thereby, it is possible to realize calculation speeding up and formulation as a constraint satisfaction problem.

  Further, the model creation means calculates an average value, a minimum value, and a maximum value of work time excluding zero-hour work for each scheduling model as a rounding method of the work period, and an optimum range based on the calculation result. It is preferable to round the construction period. As a result, the construction period is greatly rounded, so that information on the construction period difference is not lost. In addition, by rounding the work period finely, the work period can be rounded with an optimum width that does not spend a great deal of calculation time.

  Further, the storage means includes problem unit setting information for specifying a process or a group of processes as a target of a scheduling model, problem setting information for setting what optimization problem to solve the set problem unit, Problem information for setting information necessary for a set problem, and the model creating means creates a scheduling model based on the problem unit setting information, the problem setting information, and the problem information, The schedule planning means preferably optimizes the schedule for each scheduling model obtained by the model creating means, and sets a delivery date for the immediately upstream scheduling model based on the calculated schedule result. Thereby, accurate schedule creation and man-hour calculation in a practical time can be realized. In addition, by setting only the bottleneck process as a precise optimization problem, high-speed calculation can be realized. Furthermore, if the process downstream from the bottleneck process is set as a precise optimization problem, the post-request date can be optimized under accurate conditions.

  The present invention also relates to a scheduling method for performing production scheduling of a production object consisting of a plurality of processes, process connection information for setting a connection order relationship of the processes accumulated in advance, and each block included in the process Block flow information for setting the movement route, work work period information for setting the work period in each process of each block, and a read step for reading the constraint conditions of each process, and the process from the information obtained by the read step For each of the scheduling model obtained by the model creation step, the model creation step for creating a scheduling model based on the process data after the rearrangement obtained by the interpretation step, and the model creation step. A scheduling step for optimizing the schedule; And an outputting step of outputting a scheduling result obtained by the planning step. Thereby, highly accurate production scheduling can be realized. In addition, high-precision large-scale scheduling in a practical time can be realized.

  Furthermore, it is preferable to have a simulation step of performing simulation for all processes or a preset number of processes based on the scheduling result obtained by the schedule planning step. Thereby, the scheduling content can be confirmed along with the change of time, and it is possible to make it easy for the user to grasp the accuracy. Accordingly, the user or the like can accurately grasp the process of the bottleneck portion expected in the future work process, and can perform quick and reliable correction (work change) from the contents.

  Furthermore, it is preferable to have a plan data correction step for correcting the scheduling result obtained by the schedule planning step based on the simulation result obtained by the simulation step. As a result, it is possible to obtain scheduling results reflecting the know-how and the like possessed by experts who cannot be described as input conditions.

  Furthermore, it is preferable that the model creation step executes scheduling by rounding a work period of each work obtained by the interpretation step. Thereby, it is possible to realize calculation speeding up and formulation as a constraint satisfaction problem.

  Further, the model creation step calculates an average value, a minimum value, and a maximum value of work time excluding zero-hour work for each scheduling model as a rounding method of the work period, and an optimum width based on the calculation result. It is preferable to round the construction period. As a result, the construction period is greatly rounded, so that information on the construction period difference is not lost. In addition, by rounding the work period finely, the work period can be rounded with an optimum width that does not spend a great deal of calculation time.

  Further, the reading step includes problem unit setting information for specifying a process or a group of processes as a unit to be a scheduling model, problem setting information for setting an optimization problem to solve the set problem unit, Reading problem information for setting information necessary for a set problem, and the model creating step creates a scheduling model based on the problem unit setting information, the problem setting information, and the problem information, It is preferable that the schedule planning step optimizes the schedule for each scheduling model obtained by the model creation step, and sets a delivery date for the immediately upstream scheduling model based on the calculated schedule result. Thereby, accurate schedule creation and man-hour calculation in a practical time can be realized. In addition, by setting only the bottleneck process as a precise optimization problem, high-speed calculation can be realized. Furthermore, if the process downstream from the bottleneck process is set as a precise optimization problem, the post-request date can be optimized under accurate conditions.

  The present invention can be provided by a scheduling program for causing a computer to execute the scheduling method according to any one of claims 7 to 12. Thereby, highly accurate production scheduling can be performed. Moreover, the scheduling process in the present invention can be easily realized by installing the program.

  Furthermore, the present invention can be provided by a computer-readable recording medium on which the program according to claim 13 is recorded. Accordingly, the scheduling program can be easily installed on a plurality of other computers using the recording medium. Moreover, the scheduling process in the present invention can be easily realized by installing the scheduling program.

  According to the present invention, highly accurate large-scale scheduling in a practical time can be realized.

<Outline of the present invention>
The present invention does not model and schedule the entire target process group at once, but considers the importance of each process or process group and models individual process units or processes of two or more processes as one process. Modeling is done as a group, and the whole is modeled as a collection of subproblems. In addition, by implementing scheduling in a direction that goes back from the delivery date, compliance with the delivery date is realized. Therefore, the delivery time is instructed by giving the start time as a request time for the upstream process from the downstream process. Furthermore, process scheduling is realized from various viewpoints by setting various problems such as one machine problem, parallel machine problem, allocation problem, and constraint satisfaction problem for each partial problem.

  As a result, for example, a large-scale process scheduling problem for a large-scale single-order product or the like is realized in a practical time. In addition, precise scheduling is possible for particularly important processes such as complicated constraints that are likely to become bottlenecks in the entire factory.

  Here, a specific example of the above-described content will be described. For example, in the production process, the outline process of the entire factory is “material input” → “processing” → “assembly 1” → “assembly 2” → “inspection / adjustment” → In the case of “Assembly 3” → “Entire assembly”, the scheduling flow is “Overall assembly” → “Assembly 3” → “Inspection / adjustment” → “Assembly 2” → “Assembly 1” → “Processing” → “Materials” To “input”. That is, a schedule is obtained for each process, with the start date of the downstream process as the delivery date of the upstream process in the direction going back from the entire assembly. As a result, for example, it is possible to perform scheduling for the first purpose of observing the delivery date.

<Embodiment>
Hereinafter, a preferred embodiment of a scheduling device, a scheduling method, a scheduling program, and a recording medium having the program recorded thereon having the above-described features will be described in detail with reference to the drawings. In the following description, a large ship such as a tanker will be used as the production object. In addition, it is assumed that the outline process of the whole factory is “input” → “processing” → “small assembly” → “assembly” → “equipment / painting” → “formation” → “installation”. In addition, about the production target in this invention and the general process of the whole factory, it is not limited to this.

  In the following description, “process” refers to a place where manufacturing is performed using a specific resource or a certain stage of manufacturing. “Work” refers to a part of the manufacturing process that can be performed using a specific resource in the process. In addition, “block” indicates materials, parts, and the like that are targets of manufacturing work in the process.

<Scheduling device: hardware configuration>
Here, a general-purpose personal computer, a server, or the like can be used as the scheduling device in the present invention. In addition, scheduling processing can be realized by installing an execution program for realizing the present invention in a general-purpose PC.

  FIG. 1 is a diagram illustrating an example of a schematic configuration of hardware in a scheduling apparatus. 1 includes an input device 11, an output device 12, a drive device 13, an auxiliary storage device 14, a memory device 15, a CPU (Central Processing Unit) 16 that performs various controls, and a network connection. These devices are connected to each other via a bus B in a state where data can be transmitted and received between them.

  The input device 11 has a pointing device such as a keyboard and a mouse operated by the user, and inputs various operation signals such as a program execution instruction from the user. The output device 12 includes a display for displaying various windows and data necessary for operating the computer main body for performing the processing according to the present invention, a speaker for outputting sound, and the like. And results can be displayed. Furthermore, the output device 12 may have a function such as a printer. In that case, for example, various information that can be acquired such as a scheduling result is printed on a print medium such as paper and provided to a user or the like. be able to.

  Here, in the present invention, the execution program installed in the computer main body is provided by, for example, the recording medium 18 such as a CD-ROM. The recording medium 18 on which the program is recorded can be set in the drive device 13, and the execution program included in the recording medium 18 is installed from the recording medium 18 to the auxiliary storage device 14 via the drive device 13.

  Further, the drive device 13 can record the execution program according to the present invention on the recording medium 18. Thus, the recording medium 18 can be used for easy installation on a plurality of other computers, and information processing in the scheduling process of the present invention can be easily realized.

  The auxiliary storage device 14 is a storage means such as a hard disk, and can store an execution program according to the present invention, a control program provided in a computer, and the like, and can perform input / output as necessary. The memory device 15 includes a ROM (Read Only Memory), a RAM (Random Access Memory), and the like, and stores various data necessary for processing in the present invention.

  Based on a control program such as an OS (Operating System) and an execution program read and stored by the memory device 15, the CPU 16 performs various operations and data input / output with each hardware component, etc. Information processing in scheduling processing can be realized by controlling processing. Further, the CPU 66 can acquire various information necessary during the execution of the program from the auxiliary storage device 14, and can store the processing result in the auxiliary storage device 14.

  The network connection device 17 can be obtained by connecting an execution program from another terminal connected to the communication network or executing the program by connecting to a communication network or the like via a LAN (Local Area Network) cable or the like. The execution result obtained or the execution program in the present invention can be provided to other terminals.

  With the hardware configuration as described above, the scheduling process according to the present invention can be realized at a low cost without requiring a special device configuration.

<Scheduling device: functional configuration>
Next, a functional configuration example in the scheduling apparatus will be described with reference to the drawings.

<Example 1>
FIG. 2 is a diagram illustrating an example of a functional configuration of the scheduling apparatus according to the first embodiment. The scheduling apparatus 10 shown in FIG. 2 includes an input means 21, an input interpretation means 22, a model creation means 23, a schedule plan creation means 24, a database 25, an output means 26, a data correction means 27, and a control means. 28.

  The input means 21 has a pointing device such as a keyboard and a mouse operated by a user (such as a user) who uses the computer, and inputs various operation signals such as execution of a program from the user. In addition, as input data for performing scheduling in the present invention, for example, process connection information for setting the connection order relationship of processes, block flow information for setting the movement path of each block included in the process, There are constraints such as work work period information for setting a work period in each process of the block and information on the number of workers for setting the number of workers in each process.

  Specifically, overall process path pattern data (flow pattern table), block-specific quantity data (quantity table), block-specific flow pattern data (branch connection table), process data (processing speed table), transfer time data, mounting order It consists of data (mounting block input table), mounting start date (mounting block input table), process data (process conditions), and the like.

  Note that the above-described process data includes process-specific work conditions, optimization patterns, and the like. For example, as process conditions, for each process ID, building, line, line classification, and line name, a serial processing flag (a plurality of flags is used depending on a flag Are processed in series.), The number of platens, the construction period (the unit is, for example, time, etc., and one day is 8 hours), JIT (Just In Time) penalty, and the like.

  In addition, the leveling restriction conditions in the previous process can be input. The pre-process leveling condition is composed of continuous days, minimum number of blocks, maximum number of blocks, and the like. Here, the restriction of leveling the previous process is, for example, setting such that “the number of blocks whose previous process is Y is always W or more and Z or less for a certain consecutive X days”. Say. That is, for each line (Y) in the previous process, the number of consecutive days (X: standardized evaluation target period), the minimum number of blocks (W), the maximum number of blocks (Z), and the penalty value are designated.

  The JIT penalty is JIT penalty pattern data that can be described in a predetermined function pattern based on the number of days elapsed and can be set for each work. Further, as a restriction of JIT, for example, “a block X must not be completed before the Z date before the post-process (stage) request date Y of the block. Also, it must not be completed later than Y. Line In addition, it is possible to set “specify five penalty patterns”.

  In addition to the constraints described above, for example, constraints in the assembly process can be input. Examples of the constraints in this case include “construction pitch, discontinuous, continuous constraints, and production period constraints”. Etc. can be set. In addition, as the above-described non-continuous constraint, for example, “in two consecutive days in line X, a specified series name (a group of blocks having almost the same specifications is set as a series), block name, two lot names (1 In one process, a restriction such as “does not make a combination of two blocks to be processed at the same time on the same surface plate“ consecutively two blocks per lot ”” can be set. Further, as a continuous constraint, for example, a constraint such as “continue a specified series name, block name, and two lot names on two consecutive days in line X” can be set. Further, as a production period constraint, for example, “In the line X, the series name Z is produced at a constant pitch during the designated period. In the line X, the series name W is produced within the designated period. Constraints such as “start date, end date, penalty level” can be set for each line, line, and series. Note that the start date and the end date are days with respect to the construction pitch, and the number of blocks in each series can be acquired from the block-specific flow data.

  Further, as another constraint condition that can be input by the input means 21, for example, as a constraint of the block fixed construction building (the processing line of the block X is always Y), for example, in the scheduling and simulation, the block flow over the entire factory is It is possible to set a constraint such as “the flow pattern data by block (branch connection table) is necessary because it is necessary”. In addition, as a restriction of the preceding order, for example, “When crossing processes in an assembly, the preceding process must be completed within Y days or more and Z days or less in the start date of the succeeding process. Depends on the data (branch join table). You can set a constraint such as “Specify five penalty patterns for each line.” In addition, there are restrictions on the leveling of work in the entire assembly process, restrictions on the assembly make span of the same ship, and the like. The block-specific flow pattern described above will be described later.

  As described above, various types of information input by the input unit 21 can also be acquired in a text file format from an external device or the like connected via a communication network. These files are stored in the database 25.

  The input interpreter 22 reads predetermined data necessary for scheduling from the various information described above from the database 25 and performs an error check on the input data. The input interpreter 22 converts the data into a predetermined data structure necessary for creating a scheduling model. Specifically, the input interpretation unit 22 rearranges the input processes in the order of going back from the downstream to the upstream. By using this data to create models and schedules, it is possible to perform scheduling for the first purpose of complying with delivery dates. If there is an error in the input data as a result of the error check, the input interpretation unit 22 can cause the output unit 26 to output a message to that effect and prompt the user to input again.

  In addition, the model creating unit 23 is configured for each subproblem consisting of at least one process (work place) or a group of processes based on the data rearranged from the downstream to the upstream created by the input interpreting unit 22. Create a scheduling model according to the problem description format.

  Note that the model creation means 23 performs scheduling by rounding the work period of each work input to the database 25. As a result, the calculation can be speeded up and formulated as a constraint satisfaction problem.

  The model creation means 23 calculates an average value, a minimum value, and a maximum value of the work time excluding the zero-hour work for each scheduling model as a rounding method of the work period, and an optimum width (for example, based on the calculation result) Round the work period in units of time such as 15 minutes). As a result, the construction period is rounded with an optimum width so that the construction period will not be lost due to a large rounding of the construction period, and the calculation time will not be consumed enormously by rounding the construction period finely. be able to.

  Further, the model creating means 23 sets problem unit setting information for specifying a process or a group of processes as a target of the scheduling model stored in the database 25 and the like, and what optimization problem to solve the set problem unit. One or more scheduling models are created using the problem setting information to be set and the problem information for setting information necessary for each set problem. Thereby, accurate schedule creation and man-hour calculation in a practical time can be realized. In addition, by setting only the bottleneck process as a precise optimization problem, high-speed calculation can be realized. Furthermore, if the process downstream from the bottleneck process is set as a precise optimization problem, the optimization can be performed under the condition that the request date for deferral is accurate.

  The schedule planning means 24 has a solution section for each problem such as one machine problem, parallel machine problem, allocation problem, constraint satisfaction problem, etc., and optimizes the schedule for each scheduling model created by the model creating means 23. Based on the calculated schedule result, a delivery date for the immediately upstream scheduling model is set. In other words, the schedule planning means 24 performs scheduling so that, for example, the work start date of each process becomes a predetermined date with respect to the request date for the postponement of each process.

  Here, the one machine problem is a case where the process work is performed by only one machine (one work place). As a condition as one machine problem, for example, “(a) In one process, there is one work place (surface plate), and only one block can be processed.”, “(B) Processing time is different.The processing time for each block by process is calculated by the quantity / processing speed. ”,“ (C) Delivery time required by each block or downstream process ”.

  The parallel machine problem is intended for, for example, the formation process, and in these processes, a plurality of blocks can be processed in one process, and the number of platens is designated for each process. In addition, each process of formation has multiple surface plates and is an important part in terms of scheduling between “assembly” and “mounting”, so each process is formulated individually as a parallel machine problem To do.

  In other words, the conditions that should be considered as a parallel machine problem are: “(a) A plurality of blocks can be processed simultaneously in parallel at a plurality of work places (surface plates) in one step”, “(b) Each block. The processing time is different, and the processing time for each process block is calculated by the quantity / processing speed. ”“ (C) Each block has a delivery date required by the downstream process ”.

  In addition, the constraint satisfaction problem is not only general and simple constraints such as multiple simultaneous processing and delivery time, but also a fixed pitch input, load leveling constraints, and a prior order relationship when you want to schedule multiple processes at once. For example, a formulation for a problem with various complicated constraints for precise scheduling is performed. In the present embodiment, the entire assembly process is modeled by the constraint satisfaction problem for the assembly process having the most important and many constraint conditions.

  That is, as a main condition of the constraint satisfaction problem, “(a) In one process, a plurality of blocks can be processed in parallel at a plurality of work places (surface plate)”, “(b) Each block. Processing time is constant (regardless of block size), but processing time is not constant, and there are also mixed time processing steps determined by the quantity / processing speed. ”,“ (C) Block to process The throwing time interval is constant.Putting blocks into each process is thrown at a constant pitch calculated by the work time / number of plates. However, if the processing time is not constant, the throwing pitch is also not constant. " , “(D) Each block has a delivery date required by the downstream process”.

  Moreover, the schedule planning means 24 performs scheduling based on the constraint conditions such as (A) to (H) shown below.

“(A) Block processing process and process order are fixed (each block is processed by block flow designation and process order is fixed. In addition, when multiple processes are crossed in an assembly process, "(B) Combining a plurality of blocks (when a plurality of blocks are combined into one block, the process processing can be started only after all the combined blocks are aligned.)" “(C) Fixed days, existence of fixed schedule (there is a number of days (building pitch) for processing a block group for one ship uniformly for each ship in all processes.) In addition, a relative manufacturing period is set for each block series, and there may be a series having a fixed schedule with a fixed pitch within the manufacturing period.) ”,“ (D) Continuity Blocks you do not want to have (There are combinations of blocks that you do not want to process continuously for each process.), "(E) Blocks that you want to continue (There are combinations of blocks that you want to process continuously for each process.) .) ”,“ (F) Existence of 2 lots / lot (There is a combination of 2 blocks to be processed simultaneously on the same surface plate for each process. In addition, process processing starts only when the two selected blocks are aligned. ”,“ (G) Upstream process (small assembly process) leveling (Assuming start date of assembly process as delivery date of upstream process (small assembly), each block is allocated to each process of upstream process. ” At this time, there are upper and lower limits of the number of blocks for each process in the upstream process. ”,“ (H) Blocks calculated by deadline compliance (downstream process (equipment / painting, formation) ” The different request date is set in the assembly process. Compliance as deadlines distinguish the final step. In the series process in the assembly process, in the prior step to comply with the process start date of succeeding process as delivery.). "
The schedule planning means 24 performs scheduling based on the various information described above, and the result is output by the output means 26.

  The database 25 is a storage means for storing various data input by the input means 21, various models such as models created by the model creation means 23, scheduling results created by the schedule planning means 24, and the like. Data can be input and output as needed. The database 25 stores the above-described problem unit setting information, problem setting information, problem information, and the like. A specific data configuration example stored in the database 25 will be described later.

  The output means 26 outputs the scheduling result such as a screen display such as a Gantt chart, the scheduling processing result such as the scheduling result and the operation rate in a file format or the like. Specifically, the output unit 26 includes a display for displaying the scheduling result, a speaker for outputting sound, and the like. The output unit 26 may have a printing function such as a printer, and may output various information that can be acquired to a printing medium. The output unit 26 outputs an error message or the like when an error is found in the input data check or the like in the input interpretation unit 22.

  The data correction means 27 corrects input data when it is determined that a change is necessary based on the scheduling result created by the schedule planning means 24. Note that the data correction means 27 may set a reference parameter or the like in advance and compare the parameter with the scheduling result to change the input data of an item that needs correction to a predetermined value. In addition, an expert or the like may correct the input data with reference to the scheduling result displayed on the output means 26.

  The control means 28 controls the entire components in the scheduling device 10. Specifically, the control unit 28 reads predetermined data from the beta base, for example, in response to a scheduling execution instruction from the input unit 21, and performs control such as analysis of input data, model creation, schedule planning, and the like. To do.

<Database: Data structure>
Here, the database 25 described above will be described with reference to the drawings.

  FIG. 3 is a diagram illustrating an example of process data in which process connection relationships are set. 3A shows the flow data indicating the process relationship, and FIG. 3B shows the contents of the flow data corresponding to FIG. 3A. Moreover, in the example shown in FIG. 3, it has each process of process 1-8.

  The first digit of the table in FIG. 3B indicates “Process No.”. In addition, when "process No." is 0, it represents the population process in which the thing is thrown in. The second and subsequent digits indicate “connection destination process No. (post-process No.)”. However, own process No. Is described, it is possible to repeat (loop) the number of times described. For example, the process No. in FIG. In the case of 7, this indicates that the process can be repeated twice.

  FIG. 4 is a diagram illustrating an example of process processing speed data in which the processing capability of each process is set. As an item included in the process processing speed data, for example, process No. ”,“ Quantity ”,“ processing type ”,“ constant work period ”, and the like. In the example shown in FIG.

  The process processing speed data shown in FIG. 4 stores basic data for calculating the processing time in each process. Specifically, for example, in FIG. 4, when a production target having a quantity 1 = 10 is input to the process 1, the processing time is 10/1 = 10 hours. That is, the work time (process time) is determined by the quantity / quantity processing speed.

  The “processing type” is a type in which a preset type is accumulated. For example, when the “processing type” is 1, the production object has a plurality of quantities, and the corresponding processing speed is set. If it has, the maximum processing time is defined as the processing time in the process. When the “processing type” is 2, when the production object has a plurality of quantities and has a corresponding processing speed, the sum of the processing times is set as the processing time in the process. Further, the “certain construction period” is a construction period that requires the process. If a fixed work period is designated, the quantity is ignored and the designated fixed work period is adopted.

  FIG. 5 is a diagram illustrating an example of the quantity data of the production target in which the processing target quantity of the production target is set. The item data item shown in FIG. 5 includes “object No.” and each object quantity (1 to N) data number corresponding to each object. In the example shown in FIG. 5, the number of production objects is P.

  FIG. 6 is a diagram illustrating an example of process flow data of a production target in which a process for processing the production target is set. FIG. 6 shows the quantity data of the production object corresponding to each quantity. Specifically, FIG. 6A shows an example of process flow data when “Production object No.” is 1, and FIG. 6B shows the case where “Production object No.” is 2. An example of the process flow data is shown. Further, the process flow data of the production object shown in FIG. 6 indicates “production object No.” in the first line, and “process No.” and “connection destination process No. ( Post-process No.) ”. Here, the production objects referred to in the “production object No.” on the first line are the object Nos. 1 to P shown in FIG. Corresponding to

  Further, in the “process No.” in the second and subsequent lines of the process flow data, when “process No.” is 0, it represents a population process in which an object is input. Further, in the “connection destination process No. (post-process No.)”, the own process No. Can be repeated up to the number of times (loop). In the last digit of the process flow data, the production target No. Is accumulated.

  FIG. 7 is a diagram illustrating an example of process resource constraint data in which the number of process resources is set. The process resource constraint data items shown in FIG. 7 include “process No.”, “number of machines”, “number of personnel”, “number of additional personnel”, “number of required personnel per unit quantity”, and the like.

  For example, when “Process No.” is 2, it indicates that the number of machines = 5, that is, the number of production objects that can be simultaneously processed is five. In addition, the limit of the required number of people calculated from the number of people required per unit quantity is 8 people, indicating that the maximum number of people that can be added when it cannot be handled by any means is 2 people. . However, the number of personnel and the number of people that can be added are limited to the maximum total.

  FIG. 8 is a diagram showing an example of delivery date data in which the delivery date of the product is set. The items of the delivery date data shown in FIG. 8 are configured to have, for example, “final product No.” and “delivery date (year / month / day)”. The delivery date data shown in FIG. 8 sets delivery dates for a plurality of final products to be produced.

  FIG. 9 is a diagram illustrating an example of problem setting data in which a partial problem is set. The problem setting data shown in FIG. 9 is configured to have “process No.” and “problem code”. The problem setting data shown in FIG. 9 is for setting that the whole factory is divided into partial problems.

  For example, in the example shown in FIG. 9, when the “problem code” is 1, a process (process in which a delivery date is specified) that forms the final product is set. When the “problem code” is 2, a parallel machine scheduling problem (including one machine problem) is set. This is a scheduling for each process alone.

  If “problem code” is 3, an assignment problem is set. This is a scheduling for each process alone. If the “problem code” is 4- * (* is an integer), a constraint satisfaction problem is set. This is a scheduling in which a plurality of processes having the same code are grouped. In the present invention, scheduling is performed based on the data example as described above.

<Scheduling Procedure in Embodiment 1>
Next, the scheduling procedure in the first embodiment will be described with reference to a flowchart. FIG. 10 is a flowchart illustrating an example of a scheduling procedure according to the first embodiment. In the flowchart shown in FIG. 10, first, each item of the database described above is read (S01), input data is checked for errors, and the input data is converted into a predetermined data structure (S02). Next, based on the data interpreted in S02, a scheduling model as described above is created in accordance with the problem description format for each partial problem as shown in FIG. 9 (S03).

  Next, scheduling is performed for each scheduling model obtained in S03 for various problems such as one machine problem, parallel machine problem, allocation problem, constraint satisfaction problem, and the like (S04). Next, the created schedule result is displayed and output on the screen (S05), and the result is also output to the database 25 (S06).

  Here, it is determined whether or not the result is OK (S07). Whether or not the result is OK may be determined by comparing the preset parameter with the output result as described above, or refer to the scheduling result displayed by an expert or the like. Then, it may be determined whether or not it is OK. Further, the same determination as described above may be performed using the result output to the database 25.

  Next, when the result is not OK (NO in S07), the input data in the database is corrected (S08), the process returns to S01, and the processes after the input data reading process described above are performed. If the result is OK in the processing of S07 (YES in S07), the processing is terminated.

<Output example>
Here, the output contents will be described with reference to the drawings.

<Gantt chart>
FIG. 11 is a diagram illustrating an output example of a scheduling result in the present invention. The output example shown in FIG. 11 shows the output result by the Gantt chart. In FIG. 11, the schedule result (processing status) in one process is shown.

  Here, FIG. 11 is a graph showing the processing flow of one production object with the vertical axis indicating the machine name of this process and the horizontal axis indicating time. Become. Specifically, the Gantt chart display screen 30 shown in FIG. 11 is configured to include a schedule target display area 31, a button selection area 32, a Gantt chart display area 33, and a scroll bar 34.

  The schedule target display area 31 displays the schedule calculation execution time, the number of jobs in a predetermined process, and the number of lines (machines). The button selection area 32 displays buttons for performing enlargement / reduction of the Gantt chart, changing settings, printing the Gantt chart, and the like.

  The Gantt chart display area 33 displays the schedule and size (load) of each block for each line. The Gantt chart can be displayed for a predetermined period by the scroll bar 34.

  Note that the Gantt chart according to the present invention may be a Gantt chart other than that shown in FIG. In addition to the Gantt chart, an important curve, a process work waiting block number graph, and the like can be displayed.

  In addition, each work on the line displayed in the Gantt chart display area 33 is displayed in a color-coded manner for each area, or is displayed using diagonal lines or mesh lines in order to be distinguished from other works. Moreover, you may identify the delivery date compliance degree of each operation | work by highlighting, such as a color and blinking. Thereby, the user etc. can grasp | ascertain the content for every process correctly from the displayed content, for example.

<Flow data by block>
FIG. 12 is a diagram illustrating an example of block-specific flow data. The block-specific flow data shown in FIG. 12 shows the relationship of each block in each of the blocks B1 to B420 as an example. As shown in FIG. 12, for example, in a work process in shipbuilding, the blocks output in the processing procedure of “processing” → “assembly” → “formation” → “loading” → “quay” include branching and merging, A plurality of blocks are combined into one block.

  Note that branching means that, after completion of a process operation with one block, when parallel operations in a plurality of downstream processes are required, a duplicate having the same data is created and sent to a plurality of downstream processes. In addition, “merging” means that after one block is branched and divided into a plurality of blocks, the block becomes one block again in one process on the downstream side. Furthermore, it means that a plurality of blocks different from a combination are aggregated into one block in a process having a plurality of blocks.

<Example of database output>
Next, an example of a scheduling result output to the database in the scheduling process described above will be described with reference to the drawings.

  FIG. 13 is a diagram illustrating an example of a statistical output related to a processing time representing an operation status of a process. The statistical output items relating to the processing time shown in FIG. 13 include, for example, “process No.”, “number of processes”, “processing time”, “start time”, “end time”, “operation rate”, and the like. Has been. The processing time indicates “average”, “minimum”, “maximum”, and “σ (standard deviation)” values based on the number of processes in each process. Therefore, the operation status for each process can be grasped numerically from FIG.

  Moreover, FIG. 14 is a figure which shows an example of the statistical output regarding the used resource showing the operation condition of a process. 14 includes, for example, “process No.”, “number of machines used”, “number of employees”, “man-hours”, and the like. Further, the average, minimum, maximum, and σ (standard deviation) values are output for “number of machines used” and “number of employees”. Thereby, the usage status of resources can be grasped by numerical values. 14 is used as a production cost in man-hour calculation.

  According to the embodiment described above, highly accurate production scheduling can be realized. Specifically, instead of modeling and scheduling the entire target process group at once, considering the importance of each process or process group, modeling individual process units or processes with two or more processes By modeling as one group and modeling the whole as a collection of partial problems, a large-scale scheduling problem can be processed in a shorter time (practical time) than before. In addition, it is possible to automate the scheduling that relies on human hands. Furthermore, it becomes possible to incorporate a huge amount of constraints that are difficult to consider manually.

<Example 2>
Next, a second embodiment (embodiment 2) according to the present invention will be described. In the second embodiment, by performing a simulation based on the scheduling result obtained by the plan schedule creation unit 24, the scheduling content can be confirmed with a change in time, and the accuracy of the user can be easily grasped. be able to. Here, a functional configuration example of the scheduling apparatus 10 for realizing the second embodiment will be described with reference to the drawings.

  FIG. 15 is a diagram illustrating an example of a functional configuration of the scheduling apparatus according to the second embodiment. In FIG. 15, the same reference numerals are used for the same configuration as the functional configuration of the first embodiment shown in FIG. 2, and the description thereof is omitted.

  The scheduling apparatus 10 illustrated in FIG. 15 includes a simulation unit 41 in addition to the functional configuration illustrated in the first embodiment. Based on the scheduling result, the simulation unit 41 performs a simulation for every process or for a preset number of processes. Specifically, the simulation unit 41 performs simulation by modeling various resources such as processes and production objects using a discrete system simulation method or the like.

  Here, the discrete system simulation generally refers to a simulation performed based on an event-driven driving principle, and various types of software corresponding to the simulation level are commercially available. Therefore, in this embodiment, such commercially available software may be used, or self-made software may be used.

  Further, the simulation unit 41 outputs the simulation result to the database 25 and the output unit 26. As a result, it is possible to perform file output related to the system status such as animation display on the screen, detailed event information, and staying status.

Next, a scheduling process procedure in the second embodiment will be described with reference to the drawings. FIG. 16 is a flowchart illustrating an example of a scheduling process procedure according to the second embodiment.
In addition, since the process sequence from S11 to S18 is the same as that of S01 to S08 described above, description thereof is omitted here.

  Here, in the scheduling process shown in FIG. 16, when the result is OK in the process of S17 (YES in S17), next, a simulation is performed based on the scheduling result (S19). Simulation is performed by modeling various resources such as processes and production objects using a simulation technique. As a result, file output relating to the system status such as animation display on the screen, detailed event information, and staying status is performed.

  Further, the simulation result obtained by the process of S19 is displayed on the screen (S20) and is also output to the database 25 (S21). Next, it is determined whether or not the simulation result is OK (S22). Note that whether or not the simulation result is OK is determined based on the preset parameter as described above, or the displayed result. Is judged by referring to the experts.

  Here, when the simulation result is not OK (NO in S22), the input data is corrected (S23), and the process returns to S11 to perform the process after the input data reading process. Moreover, in the process of S22, when the result is OK (YES in S22), the process ends.

  Thereby, highly accurate scheduling can be performed. The simulation program displays an animation so that the progress of the work on the screen and the staying status between the processes can be visually confirmed. Also, a Gantt chart as shown in FIG. 11 by simulation can be output, and various statistical outputs as shown in FIG. 13 by simulation are also possible.

  In performing the simulation process as described above, the schedule planning means 24 prepares data for simulation. FIG. 17 is a diagram illustrating an example of processing order data output for simulation. 17 includes, for example, “time”, “process No.”, “production object No.”, “final product No.”, and the like from the start of the process. Has been.

  Note that FIG. 17 outputs the processing start time in the process of each production object among the scheduling results. Based on this output, the simulation program determines the processing start time of the production object in each process. Moreover, in FIG. 17, it can output also in Example 1 as a processing order of the production target object in each process.

<Simulation result: Animation display example>
Next, an animation display example will be described with reference to the drawings as a simulation result in the second embodiment. FIG. 18 is a diagram illustrating an example of a simulation result in the second embodiment. Note that the animation screen 50 by simulation in FIG. 18 displays the work status at each work place at the time displayed in the time display area 51.

  For example, in the ship building process shown in FIG. 18, it can be roughly divided into a “discovery” area, a processing area, a small assembly area, an assembly area, a formation / fitting area, a loading (dock) area, a quay area, and the like. It is displayed. In addition, among the displayed areas, each time the time in the time display area elapses, the working area blinks, or is identified and highlighted by color coding or diagonal lines.

  Further, the animation screen 50 shown in FIG. 18 has a block number display area 52, and the block number display area 52 displays, for example, the number of blocks waiting for work, waiting for work, and waiting to be carried out at that time. Also, the time displayed in the time display area 51 can be advanced faster than the actual time by the set speed information.

  Thereby, the scheduling content can be confirmed along with the change of time, and it is possible to make it easy for the user to grasp the accuracy. Accordingly, the user or the like can accurately grasp the process of the bottleneck portion expected in the future work process, and can perform quick and reliable correction (work change) from the contents.

<Example 3>
Next, Example 3 will be described. In the third embodiment, not only the input data is corrected from the simulation result as shown in the second embodiment, but also the function of directly correcting the output scheduling result is provided. Here, a functional configuration example of the scheduling apparatus 10 according to the third embodiment will be described with reference to the drawings. FIG. 19 is an example of a functional configuration example of the scheduling apparatus according to the third embodiment. The scheduling apparatus 10 shown in FIG. 19 has plan data correction means 42 in addition to the functional configuration shown in the second embodiment.

  The plan data correction means 42, for example, refers to the simulation result and, when it is determined that the scheduling result can be directly changed, the plan data correction means 42 does not correct the input data but uses the scheduling data of the scheduling result. Correct it directly and run the simulation again. Thereby, it is possible to obtain a scheduling result reflecting the know-how possessed by an expert that cannot be described as an input condition with respect to the schedule obtained by the schedule planning means 24.

  Here, the scheduling procedure in the third embodiment will be described with reference to a flowchart. FIG. 20 is a diagram illustrating an example of a scheduling process procedure according to the third embodiment. In FIG. 20, the processing of S31 to S41 is the same processing as S11 to S21 in the above-described second embodiment, and thus description thereof is omitted here.

  Here, in the scheduling process shown in FIG. 20, it is determined whether or not the simulation result is OK in the process of S42 (S42). If the result is not OK (NO in S42), is the input data corrected? Alternatively, it is determined whether to correct the scheduling result (plan data). For example, in the process shown in FIG. 20, it is determined whether or not the input data is to be corrected (S43). If the input data is not to be corrected (NO in S43), the plan data is corrected (S44), and the process of S39 is performed. Return to, and perform the processing after the simulation processing.

  If the input data is to be corrected (YES in S43), the input data is corrected (S45), and the process returns to S31 to perform the processes after the input data reading process. Further, in the process of S42, if the result is OK (YES in S42), the process ends.

  As described above, according to the third embodiment, it is possible to obtain a scheduling result reflecting the know-how and the like possessed by experts who cannot be described as input conditions.

  As described above, according to the present invention, highly accurate scheduling can be performed. Specifically, instead of modeling and scheduling the entire target process group at once, considering the importance of each process or process group, modeling individual process units or processes with two or more processes By modeling as one group and modeling the whole as a collection of partial problems, a large-scale scheduling problem can be processed in a shorter time (practical time) than before. In addition, it is possible to automate the scheduling that relies on human hands. Furthermore, it becomes possible to incorporate a huge amount of constraints that are difficult to consider manually.

  The above-described invention can be applied to general planning problems such as high-mix low-volume production scheduling, development projects, and construction plans.

  The preferred embodiments of the present invention have been described in detail above, but the present invention is not limited to such specific embodiments, and various modifications, within the scope of the gist of the present invention described in the claims, It can be changed.

It is a figure which shows an example of schematic structure of the hardware in a scheduling apparatus. It is a figure which shows an example of a function structure of the scheduling apparatus in Example 1. FIG. It is a figure which shows an example of the process data in which the connection relation of the process was set. It is a figure of an example which shows the process processing speed data in which the processing capability of each process was set. It is a figure which shows an example of the quantity data of the production target in which the processing target quantity of the production target was set. It is a figure which shows an example of the process flow data of the production target in which the process of the production target was set. It is a figure which shows an example of the process resource constraint data in which the number of process resources etc. were set. It is a figure which shows an example of the delivery date data in which the delivery date of the product was set. It is a figure which shows an example of the problem setting data in which the partial problem was set. 3 is a flowchart illustrating an example of a scheduling procedure according to the first embodiment. It is a figure which shows the example of an output of the scheduling result in this invention. It is a figure which shows an example of the flow data according to block. It is a figure which shows an example of the statistical output regarding the processing time showing the operating condition of a process. It is a figure which shows an example of the statistical output regarding the use resource showing the operating condition of a process. It is a figure which shows an example of a function structure of the scheduling apparatus in Example 2. FIG. 10 is a flowchart illustrating an example of a scheduling process procedure according to the second embodiment. It is a figure which shows an example of the process order data output for simulation. It is a figure which shows an example of the simulation result in Example 2. FIG. FIG. 10 is an example of a functional configuration example of a scheduling device according to a third embodiment. FIG. 10 is a diagram illustrating an example of a scheduling process procedure according to the third embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Scheduling device 11 Input device 12 Output device 13 Drive device 14 Auxiliary storage device 15 Memory device 16 CPU
DESCRIPTION OF SYMBOLS 17 Network connection apparatus 18 Recording medium 21 Input means 22 Input interpretation means 23 Model preparation means 24 Schedule plan preparation means 25 Database 26 Output means 27 Data correction means 28 Control means 30 Gantt chart display screen 31 Schedule object display area 32 Button selection area 33 Gantt chart display area 34 Scroll bar 41 Simulation means 42 Plan data correction means 50 Animation screen 51 Time display area 52 Block number display area

Claims (14)

In a scheduling device that performs production scheduling of a production object consisting of a plurality of processes,
Process connection information for setting the connection order relationship of the process, block flow information for setting a movement path of each block included in the process, work work period information for setting a work period in each process of each block, Accumulation means in which constraint conditions for each process are accumulated;
Interpreting means for rearranging the steps from the information accumulated in the accumulating means in the order of going back downstream to upstream;
Model creation means for creating a scheduling model based on the rearranged process data obtained by the interpretation means;
Schedule planning means for optimizing the schedule for each scheduling model obtained by the model creating means;
And an output means for outputting a scheduling result obtained by the schedule planning means.   Based on the scheduling result obtained by the schedule planning means, a simulation means is provided for performing a simulation for all processes or for a preset number of processes, and outputting the result to the storage means or the output means. The scheduling apparatus according to claim 1.   3. The scheduling apparatus according to claim 2, further comprising plan data correcting means for correcting a scheduling result obtained by the schedule planning means based on a simulation result obtained by the simulation means. The model creation means includes
The scheduling apparatus according to claim 1, wherein scheduling is performed by rounding a work period of each work obtained by the interpreting unit. The model creation means includes
As the rounding method of the work period, an average value, a minimum value, and a maximum value of work hours excluding zero-hour work are calculated for each scheduling model, and the work period is rounded with an optimum width based on the calculation result. The scheduling apparatus according to claim 4. The storage means includes
Problem unit setting information for specifying a process or a group of processes as a target unit of the scheduling model, problem setting information for setting what kind of optimization problem to solve the set problem unit, and necessary for each set problem Problem information to set the information,
The model creating means creates a scheduling model based on the problem unit setting information, the problem setting information, and the problem information,
The schedule plan creation means optimizes a schedule for each scheduling model obtained by the model creation means, and sets a delivery date for the immediately upstream scheduling model based on the calculated schedule result. Item 6. The scheduling device according to any one of Items 1 to 5. In a scheduling method for scheduling production of a production object consisting of a plurality of processes,
The process connection information for setting the connection order relation of the processes accumulated in advance, the block flow information for setting the movement path of each block included in the process, and the work for setting the work period in each process of each block A reading step for reading construction period information and constraint conditions of each process;
An interpretation step of rearranging the process from the information obtained by the reading step in the order of going back downstream to upstream;
A model creation step of creating a scheduling model based on the rearranged process data obtained by the interpretation step;
A schedule creation step for optimizing the schedule for each scheduling model obtained by the model creation step;
An output step of outputting a scheduling result obtained by the schedule planning step.   8. The scheduling method according to claim 7, further comprising a simulation step of performing simulation for all processes or a preset number of processes based on the scheduling result obtained by the schedule planning step.   9. The scheduling method according to claim 8, further comprising a plan data correction step of correcting the scheduling result obtained by the schedule planning step based on the simulation result obtained by the simulation step. The model creation step includes:
The scheduling method according to claim 7, wherein scheduling is executed by rounding a work period of each work obtained by the interpretation step. The model creation step includes:
As the rounding method of the work period, an average value, a minimum value, and a maximum value of work hours excluding zero-hour work are calculated for each scheduling model, and the work period is rounded with an optimum width based on the calculation result. The scheduling method according to claim 10. The reading step includes
Problem unit setting information for specifying a process or a group of processes as a target unit of the scheduling model, problem setting information for setting what kind of optimization problem to solve the set problem unit, and necessary for each set problem Read the problem information to set the information,
The model creation step creates a scheduling model based on the problem unit setting information, the problem setting information, and the problem information,
The scheduling plan creation step optimizes a schedule for each scheduling model obtained by the model creation step, and sets a delivery date for the immediately upstream scheduling model based on the calculated schedule result. Item 12. The scheduling method according to any one of Items 7 to 11.   A scheduling program for causing a computer to execute the scheduling method according to any one of claims 7 to 12.   A computer-readable recording medium on which the program according to claim 13 is recorded.

Images