WO2022158225A1 - Production floor management system, work instruction method, and work instruction program - Google Patents

Abstract

A production floor management system (1) manages conditions on a production floor provided with production equipment for producing products, said system comprising: a first condition monitoring unit (21) for monitoring a first condition on the production floor; a first countermeasure setting unit (31) for setting a first countermeasure corresponding to the first condition; a second condition monitoring unit (22) for monitoring a second condition, which differs from the first condition, on the production floor; a second countermeasure setting unit (32) for setting a second countermeasure corresponding to the second condition; a countermeasure mediation unit (40) that performs mediation between the first countermeasure and the second countermeasure; and an instruction output unit (50) that, in accordance with said mediation, outputs instructions corresponding to the first countermeasure or the second countermeasure.

Description

Production floor management system, work instruction method and work instruction program

The present disclosure relates to a production floor management system that manages the state of a production floor equipped with production equipment that produces products, a work instruction method, and a work instruction program in the system.

Conventionally, when a problem occurs in a production device, a technique has been disclosed that outputs work instructions according to the method of coping while updating the method of coping with the problem (for example, Patent Document 1).

WO2018/142604

However, multiple problems may occur in production equipment, and with the technology disclosed in Patent Document 1, it is difficult to output optimal work instructions from multiple coping methods in such cases.

Therefore, the present disclosure provides a production floor management system and the like that can output optimal work instructions from a plurality of coping methods.

A production floor management system according to one aspect of the present disclosure is a production floor management system that manages a state on a production floor that includes production equipment that produces a product, the first state monitoring a first state on the production floor. a monitoring unit, a first countermeasure determining unit that determines a first countermeasure corresponding to the first state, a second state monitoring unit that monitors a second state different from the first state on the production floor, and a second countermeasure determination unit that determines a second countermeasure corresponding to a second state; a countermeasure mediation unit that mediates between the first countermeasure and the second countermeasure; and an instruction output unit that outputs an instruction corresponding to the countermeasure.

It should be noted that these general or specific aspects may be realized by systems, devices, methods, recording media, or computer programs, and any combination of systems, devices, methods, recording media, and computer programs. may be implemented with

According to the production floor management system and the like according to the present disclosure, it is possible to output optimal work instructions from multiple coping methods.

FIG. 1 is a diagram showing a mounting line to which a production floor management system according to an embodiment is applied. FIG. 2 is a configuration diagram showing an example of the production floor management system according to the embodiment. FIG. 3 is a diagram illustrating an example of a monitoring target of a state monitoring unit according to the embodiment; FIG. 4 is a diagram schematically showing the flow of operations of the production floor management system according to the embodiment. 5 is a flowchart illustrating an example of the operation of the state monitoring unit according to the embodiment; FIG. 6 is a flowchart illustrating an example of an operation of a countermeasure determination unit according to the embodiment; FIG. FIG. 7 is a table showing an example of a countermeasure candidate list according to the embodiment. FIG. 8 is a table showing an example of a priority countermeasure list according to the embodiment; 9 is a flowchart illustrating an example of the operation of a countermeasure mediation unit according to the embodiment; FIG. FIG. 10 is a table showing an example of a resource management table according to the embodiment. 11 is a flowchart illustrating an example of operations of an instruction output unit, an effect determination unit, and an update unit according to the embodiment; FIG. FIG. 12 is a table showing an example of an updated countermeasure candidate list according to the embodiment. FIG. 13 is a flow chart showing an example of a work instruction method according to another embodiment.

A production floor management system of the present disclosure is a production floor management system that manages the state of a production floor that includes production equipment that produces a product, the system comprising: a first state monitoring unit that monitors a first state of the production floor; a first countermeasure determination unit that determines a first countermeasure corresponding to the first state; a second state monitoring unit that monitors a second state different from the first state on the production floor; a second countermeasure determination unit that determines a corresponding second countermeasure; a countermeasure arbitration unit that arbitrates between the first countermeasure and the second countermeasure; and an instruction output unit that outputs an instruction.

According to this, a first measure and a second measure are determined for dealing with different first and second situations on the production floor, but the first measure and the second measure are arbitrated. , an instruction corresponding to the optimum first or second countermeasure can be output. In this way, according to the production floor management system of the present disclosure, it is possible to output optimum work instructions from a plurality of coping methods.

The first state monitoring unit monitors the first state and the first countermeasure determining unit determines the first countermeasure, and the second state monitoring unit monitors the second state and determines the second countermeasure. The determination of the second countermeasure by the department may be performed in parallel.

For example, the first countermeasure and the second countermeasure determined when the first state and the second state are monitored in parallel may be countermeasures that are difficult to make compatible, but in the present disclosure, the first Since the countermeasure and the second countermeasure are arbitrated, an instruction corresponding to the optimum first countermeasure or second countermeasure can be output. Further, when the first measure and the second measure are measures that can be compatible, it is possible to output instructions corresponding to the first measure and the second measure in parallel without performing arbitration, Efficient countermeasures can be taken.

Further, the countermeasure arbitration unit performs the arbitration based on the degree of problem set in the first state corresponding to the first countermeasure and the degree of problem set in the second state corresponding to the second countermeasure. you can go

According to this, it is possible to preferentially output countermeasures for high-problem situations.

Further, if there is no difference between the problem level set in the first state corresponding to the first countermeasure and the problem level set in the second state corresponding to the second countermeasure, the countermeasure arbitration unit The arbitration may be performed with priority given to the first countermeasure.

According to this, it is possible to maintain the operation of production equipment by prioritizing the first measure when there is no difference in the degree of problem.

In addition, the production floor management system determines whether the first countermeasure or the second countermeasure corresponding to the output instruction is executed based on at least one of the first state and the second state before and after the execution of the second countermeasure. An effect determination unit that determines the effect of the first countermeasure or the second countermeasure may be further provided.

According to this, by judging the effect of the implemented countermeasures, it is possible to verify whether the countermeasures determined in response to the conditions on the production floor were appropriate, etc., which can be used to improve future countermeasures. can be done.

Also, the first state may be the production state of the production equipment, and the second state may be the state of production resources managed on the production floor and used for production.

For example, MTTR (Mean Time To Repair) can be improved by monitoring the production status of production equipment and outputting a first countermeasure corresponding to the production status. For example, as a result of monitoring the production status, if it is detected that there is a problem in the production status (for example, if productivity or quality is poor, or if there is a defect, etc.), the first countermeasure is to solve the problem of the production status. Since this is a countermeasure for solving the problem, the MTTR can be improved by taking a countermeasure according to the first countermeasure.

Also, for example, by monitoring the state of production resources and outputting a second countermeasure corresponding to the state of production resources, MTBF (Mean Time Between Failure) can be improved. For example, if it is detected that there is a problem in the state of the resource as a result of monitoring the state of the resource (for example, if a production device or facility element is degraded, or if there is an error in the worker's work, etc.), The second measure is to solve the problem of production resource status. For example, if the problem of the status of production resources is left unsolved, a problem may occur in the production status. Therefore, the MTBF can be improved by taking measures according to the second measure. In other words, it is possible to prevent the occurrence of production problems.

Further, the countermeasure arbitration unit may perform the arbitration based on the presence or absence of the production resources required for the first countermeasure or the second countermeasure.

For example, the first measure to solve the problem of production status and the second measure to solve the problem of production resource status are often measures related to production resources, and the production resources are already in use. In this case, even if an instruction corresponding to the first countermeasure or the second countermeasure is output, it is difficult to deal with the output instruction. On the other hand, in this aspect, since arbitration is performed based on the presence or absence of production resources, it is possible to output an instruction corresponding to the optimum first or second countermeasure depending on the presence or absence of production resources.

Further, the production floor management system further includes a first learning model for detecting a first predetermined state occurring on the production floor corresponding to the production state, and the first state monitoring unit performs the first learning model. The first predetermined state may be detected based on a model. Specifically, the first state monitoring unit collects, within a predetermined period, production error information that has occurred in the production apparatus, production volume information of the product produced by the production apparatus, and production error information produced by the production apparatus. The first predetermined state corresponding to a production index related to at least one item of quality information may be detected, and the first countermeasure determination unit may determine the first countermeasure including countermeasures for improving the production index.

Thus, by using the first learning model, it is possible to effectively detect the first predetermined state.

The production floor management system further includes a second learning model for detecting a second predetermined state occurring on the production floor corresponding to the state of the production resource, and The second predetermined state may be detected based on two learning models. Specifically, the second state monitoring unit detects the second predetermined state corresponding to an operation index related to the operating state of the production equipment included in the production resource, and the second countermeasure determination unit detects the operation The second countermeasure including countermeasures for the production equipment corresponding to the index may be determined. Specifically, the second state monitoring unit detects the second predetermined state corresponding to the work index related to the work performed by the worker included in the production resource, and the second countermeasure determination unit The second countermeasure including work of the worker corresponding to the work index may be determined.

Thus, by using the second learning model, the second predetermined state can be effectively detected.

A work instruction method of the present disclosure is a work instruction method in a production floor management system that manages the state of a production floor provided with production equipment that produces a product, the method monitors a first state on the production floor, , monitoring a second state different from the first state, determining a first countermeasure corresponding to the first state, determining a second countermeasure corresponding to the second state, and determining the first countermeasure and the Arbitrating with a second countermeasure, and outputting an instruction corresponding to the first countermeasure or the second countermeasure according to the arbitration.

According to this, it is possible to provide a work instruction method that can output the optimum work instruction from multiple coping methods.

The work instruction program of the present disclosure is a work instruction program that causes a computer to execute the work instruction method described above.

According to this, it is possible to provide a work instruction program that can output optimal work instructions from multiple coping methods.

It should be noted that the embodiments described below are all comprehensive or specific examples. Numerical values, shapes, materials, components, arrangement positions and connection forms of components, steps, order of steps, and the like shown in the following embodiments are examples, and are not intended to limit the present disclosure.

(Embodiment)
An embodiment will be described below with reference to FIGS. 1 to 12. FIG.

FIG. 1 is a diagram showing a mounting line 4 to which the production floor management system 1 according to the embodiment is applied.

As shown in FIG. 1, the mounting line 4 is equipped with a plurality of production devices that produce products.

The mounting line 4 has a function of mounting a component (electronic component) on a substrate to produce a product (for example, a mounting substrate), and has a function of supplying, delivering, and retrieving the substrate to be mounted. have.

Specifically, in the mounting line 4, the substrate supply device M1, the substrate transfer device M2, the printer M3, the mounting devices M4 and M5, the reflow device M6, and the substrate recovery device M7 are arranged in this order. connected in series. Each device from the substrate supply device M1 to the substrate recovery device M7 is connected to the management device 5 via the communication network 2. FIG.

For example, the solder printing device M3, the component mounting devices M4 and M5, and the reflow device M6 perform component mounting work for mounting components on the board conveyed along the mounting line 4. That is, the substrate supplied by the substrate supply device M1 is carried into the printer M3 via the substrate transfer device M2. The printer M3 performs a solder printing operation of screen-printing solder for joining components to the board that has been carried in.

The solder-printed boards are sequentially delivered to the mounting devices M4 and M5. The mounting apparatuses M4 and M5 perform a component mounting operation for mounting components on the substrate after solder printing.

The component mounting apparatuses M4 and M5 are equipped with a base, a board transfer section, a component supply device, a mounting head, and the like. A substrate is arranged on the base. The substrate transfer section can transfer a substrate transferred from an upstream device to a downstream device. The component supply device can supply components to the mounting head. A component supply device is provided with a plurality of tape feeders for supplying components to the mounting head. The mounting head can pick up the component from the tape feeder by suction, move it above the board, and mount the component at the mounting position on the board. The mounting head is equipped with suction nozzles that suction and hold components and that can move up and down individually. Component mounting work is performed by such component mounting apparatuses M4 and M5.

Then, the board after component mounting is carried into the reflow device M6 and heated according to a predetermined heating profile. As a result, the solder for joining components printed on the heated substrate is melted and solidified. By soldering the component to the board in this manner, a mounting board having the component mounted on the board is completed. The completed mounting substrate is recovered by the substrate recovery device M7.

Next, the configuration of the production floor system 1 will be explained using FIG.

FIG. 2 is a configuration diagram showing an example of the production floor management system 1 according to the embodiment.

The production floor management system 1 is a system that manages the status of production floors equipped with production equipment that produces products. The production floor includes, for example, the mounting line 4, inventory warehouses, preparation areas and maintenance areas. In the mounting line 4, as described above, production devices such as the mounting devices M4 and M5 and the printing device M3 and inspection devices are arranged. Materials such as parts, solder, and screen masks are stored in the inventory warehouse. In the preparation area preparation of equipment elements such as carriages, feeders, nozzles and heads is carried out. In the maintenance area, maintenance of the equipment elements, jigs, and the like is performed. The jig mentioned here is for adjusting the feeder, nozzle, head, etc., and may also be for adjusting the head moving mechanism and the transport mechanism of the equipment. Workers perform operations such as production, preparation, and maintenance in each area on the production floor, and perform operations such as transferring parts and equipment elements between areas. For example, the production floor management system 1 is a computer placed on the production floor. For example, the functions of the production floor management system 1 may be provided in the management device 5 . The production floor management system 1 may be a computer provided in one housing, or may be divided into two or more housings and implemented by two or more computers. Moreover, the production floor management system 1 may not be arranged on the production floor, and may be a computer such as a server provided outside the production floor. Note that workers are not limited to humans, and include robots, work mechanisms, and automated guided vehicles that perform the above-described work.

The production floor management system 1 is a system that outputs instructions to workers or production equipment on the production floor according to the state of the production floor. The production floor management system 1 includes an acquisition unit 10, a state monitoring unit 20, a countermeasure determination unit 30, a countermeasure mediation unit 40, an instruction output unit 50, an effect determination unit 60, an update unit 70, learning models 23, 24, 33 and 34, Also, a resource database 41 is provided. The production floor management system 1 is implemented by a computer including a processor, memory and the like. The acquisition unit 10, the state monitoring unit 20, the countermeasure determination unit 30, the countermeasure mediation unit 40, the instruction output unit 50, the effect determination unit 60, and the update unit 70 are implemented by the processor operating according to the program stored in the memory. . Moreover, the state monitoring unit 20 may be provided independently as a computer, or may be provided in the production apparatus. Moreover, the production floor management system 1 may include a plurality of state monitoring units 20 . The learning models 23, 24, 33 and 34 and the resource database 41 are stored in memory. The memory in which the programs, learning models 23, 24, 33 and 34, and resource database 41 are stored may be the same memory or different memories.

The acquisition unit 10 acquires information for monitoring the state on the production floor. For example, the acquisition unit 10 acquires information indicating the production status of the production floor. Specifically, the acquisition unit 10 acquires the results of the production process of the production equipment (specifically, productivity, quality, or the presence or absence of defects, etc.) as information indicating the production status of the production floor. The result of the production process may be a sensing history record by a sensor, or may be data input by a person. Also, for example, the acquisition unit 10 acquires information indicating the state of production resources managed on the production floor and used for production. Production resources are, for example, production equipment, facility elements, workers, materials, or jigs. For example, information indicating the state of production resources may be sensing data from a camera, sensor, or the like, or may be data input by a person. Also, for example, the acquisition unit 10 acquires event information that has changed on the production floor. More specifically, the acquisition unit 10 detects when the production apparatus stops, when the feeder, nozzle, parts, or board attached to the production apparatus is replaced, when the worker who performs the work is replaced, or when the operation data of the production apparatus is changed. Acquire information indicating that the

The state monitoring unit 20 monitors the state of the production floor through the information acquired by the acquisition unit 10 . For example, the state monitoring unit 20 detects a predetermined state based on a first condition for detecting a predetermined state (for example, a first state or a second state described later) among the states on the production floor. For example, the first condition is a learning model associated with a detection threshold corresponding to a given state. Note that the first condition may be a set detection threshold instead of the learning model.

The state monitoring section 20 has a first state monitoring section 21 and a second state monitoring section 22 . The first state monitoring section 21 and the second state monitoring section 22 each perform the operation of the state monitoring section 20 described above.

The first state monitoring unit 21 monitors the first state on the production floor. For example, the first state is the production state of the production device, and the first state monitoring unit 21 monitors the result of the production process as the production state of the production device. For example, the first state monitoring section 21 detects the first predetermined state based on the learning model 23 . The learning model 23 is a first learning model for detecting a first predetermined state occurring on the production floor corresponding to the production state of the production equipment. The learning model 23 is also a learning model (that is, the first condition) associated with the detection threshold corresponding to the production state of the production equipment. For example, the first state monitoring unit 21 monitors, within a predetermined period, at least information on production errors occurring in the production equipment, information on the amount of products produced by the production equipment, and quality information on the products produced by the production equipment. A first predetermined condition corresponding to a production index for one is detected.

The second state monitoring unit 22 monitors a second state different from the first state on the production floor. For example, the second state is the state of the production resource, and the second state monitoring section 22 monitors the state of the production resource. For example, the second state monitoring section 22 detects the second predetermined state based on the learning model 24 . Learning model 24 is a second learning model for detecting a second predetermined state occurring on the production floor corresponding to the state of production resources. Learning model 24 is also a learning model (ie, first condition) associated with a detection threshold corresponding to the state of the production resource. For example, the second state monitoring unit 22 detects a second predetermined state corresponding to an operating index related to the operating state of production equipment included in the production resource. Also, for example, the second state monitoring unit 22 detects a second predetermined state corresponding to the work index related to the work performed by the worker included in the production resource.

Here, an example of the monitoring target of the state monitoring unit 20 will be described using FIG.

FIG. 3 is a diagram showing an example of objects monitored by the state monitoring unit 20 according to the embodiment. In addition, in FIG. 3, the production apparatus is assumed to be a mounting apparatus, and the production process is assumed to be a mounting process.

As shown in FIG. 3, the state monitoring unit 20 (specifically, the first state monitoring unit 21) monitors the result of the mounting process including processes such as suction, recognition, and mounting as the production state. For example, the first state monitoring unit 21, within a predetermined period, at least information on mounting errors occurring in the mounting apparatus, information on the amount of mounted components mounted by the mounting apparatus, and information on the quality of the mounted components mounted by the mounting apparatus. A first predetermined state corresponding to one production index (specifically, the presence or absence of a mounting error (spoilage), mounting amount (productivity), mounting quality, etc.) is detected.

In addition, the state monitoring unit 20 (specifically, the second state monitoring unit 22) monitors, as the state of production resources, elements (production resources) related to the mounting process such as substrates, parts, workers, heads, nozzles, and feeders. Monitor status. For example, the second state monitoring unit 22 detects a second predetermined state (for example, deterioration of mounting devices, nozzles, and feeders, etc.) corresponding to operation indicators relating to the operating states of mounting devices included in production resources. Also, for example, the second state monitoring unit 22 detects a second predetermined state corresponding to a work index (for example, a worker's work error, etc.) related to work performed on a mounting apparatus by a worker included in the production resource. In addition to the work index, it is also related to the difference between the numerical values (length, width, thickness, coordinate positions of various marks, viscosity) on the design data of the board, solder, or parts and the numerical values actually measured by the camera or sensor. A second predetermined state may be detected that corresponds to a time index related to the measurement index, the expiry date and time of the solder or component, and the date and time of actual measurement.

Returning to the description of FIG. 2, the countermeasure determination unit 30 determines a countermeasure to be executed from among a plurality of countermeasures extracted corresponding to the state of the production floor monitored by the state monitor unit 20. The countermeasure to be executed out of the plurality of countermeasures corresponds to a first priority instruction, a second priority instruction, etc., which will be described later. For example, the countermeasure determination unit 30 determines a countermeasure to be executed based on a second condition for determining a countermeasure (for example, a first countermeasure or a second countermeasure to be described later) corresponding to a predetermined state. For example, the second condition is a learning model associated with a plurality of measures corresponding to a predetermined state and priority, including a priority set for each of the plurality of measures to be extracted. For example, the countermeasure determining unit 30 selects the first countermeasure to be executed from among the plurality of countermeasures based on the priority of each of the plurality of countermeasures extracted corresponding to the state of the production floor monitored by the state monitoring unit 20. Determine priority instructions. In other words, the countermeasure determination unit 30 determines a countermeasure (that is, the first priority instruction) to be executed from among the plurality of countermeasures based on the priority of each of the plurality of countermeasures extracted corresponding to the predetermined state. For example, in order to determine a countermeasure corresponding to a predetermined state, the countermeasure determination unit 30 uses operation information of production equipment acquired on the production floor, worker information working on the production floor, and material information used for the product. analyse. In the analysis, there are cases where countermeasures can be determined based only on the tendency of the predetermined state, and cases where the countermeasure cannot be determined based only on the tendency of the predetermined state. If countermeasures cannot be determined only by the tendency of the predetermined state, the countermeasure determination unit 30 obtains operation information of production equipment acquired on the production floor for each production resource based on event information, information on workers working on the production floor, Also, by analyzing material information used in the product, executing a predetermined countermeasure and further analyzing trends before and after the execution of the predetermined countermeasure, it is possible to determine a countermeasure corresponding to the predetermined state. Also, the analysis may be performed using an analysis unit independent of the countermeasure determination unit 30 . Further, for example, the countermeasure determination unit 30 may change the state of the production floor after the execution of the first priority instruction to the state of the production floor before the execution of the first priority instruction, which is caused by the execution of the first priority instruction. , update the priority for determining the priority for the first priority instruction (countermeasure). Also, for example, the countermeasure determination unit 30 updates the priority based on a learning model for updating the priority. The second condition may be a data table containing priorities set for each of a plurality of countermeasures instead of the learning model.

The countermeasure determination unit 30 has a first countermeasure determination unit 31 and a second countermeasure determination unit 32 . The first countermeasure determination unit 31 and the second countermeasure determination unit 32 each perform the operation of the countermeasure determination unit 30 described above.

The first countermeasure determining unit 31 determines a first countermeasure corresponding to the first state monitored by the first state monitoring unit 21. For example, the first countermeasure determining unit 31 executes a first countermeasure (that is, a first priority instruction) from among the plurality of countermeasures based on the respective priorities of the plurality of countermeasures extracted corresponding to the first state. to decide. For example, the first countermeasure determination unit 31 determines first countermeasures including countermeasures for improving the production index. Specifically, the first countermeasure determining unit 31 determines a first countermeasure for improving MTTR. MTTR is an index indicating the maintainability of a system or equipment, and the shorter the MTTR, the higher the maintainability. For example, the first countermeasure determination unit 31 determines the first countermeasure based on the learning model 33 . The learning model 33 is a learning model for determining the first countermeasure corresponding to the first state, and specifically, a learning model for updating the priority. The learning model 33 is also a learning model (that is, the second condition) associated with multiple countermeasures and priorities corresponding to the predetermined state.

The second countermeasure determining unit 32 determines a second countermeasure corresponding to the second state monitored by the second state monitoring unit 22. For example, the second countermeasure determination unit 32 executes a second countermeasure (that is, a first priority instruction) from among the plurality of countermeasures based on the respective priorities of the plurality of countermeasures extracted corresponding to the first state. to decide. For example, the second countermeasure determination unit 32 determines countermeasures (maintenance/replacement) for production equipment and facility elements corresponding to the operation index, or second countermeasures including work availability and work training for workers corresponding to the work index. to decide. Specifically, the second countermeasure determination unit 32 determines a second countermeasure for improving MTBF. MTBF is an index indicating the reliability of a system or equipment, and longer MTBF means higher reliability. For example, the second countermeasure determination unit 32 determines the second countermeasure based on the learning model 34 . The learning model 34 is a learning model for determining a second countermeasure corresponding to the second state, and more specifically, a learning model for updating the priority. The learning model 34 is also a learning model (that is, the second condition) associated with multiple countermeasures and priorities corresponding to the predetermined state. Note that the second countermeasure determined here may be completed by notifying the second countermeasure to a maintenance plan creation device or worker management device provided separately from the production floor management system 1, or the second countermeasure may be completed. 2 It may be completed after receiving the execution result of countermeasures.

The countermeasure arbitration unit 40 arbitrates between the first countermeasure and the second countermeasure. Here, the reason why mediation between the first measure and the second measure is necessary will be explained in detail with reference to FIG.

For example, as shown in FIG. 3, the result of the mounting process for picking is associated with the substrate, component, worker, head, nozzle and feeder, and the result of the mounting process for recognition is associated with the component, head and nozzle. Related and results of the packaging process with respect to mounting are related to the board, components and nozzles. For example, the first countermeasure includes countermeasures for substrates, parts, workers, heads, nozzles, or feeders, depending on the results of the production process. On the other hand, the second countermeasures also include countermeasures for substrates, parts, workers, heads, nozzles, or feeders corresponding to the state of production resources. In other words, the production resource targeted by the first measure and the production resource targeted by the second measure may overlap. If the production resources targeted by the first measure and the production resources targeted by the second measure overlap, it is difficult to implement both the first measure and the second measure at the same time. It is necessary to mediate with two countermeasures.

For example, the countermeasure arbitration unit 40 selects the first countermeasure and the second countermeasure based on the problem level set in the first state corresponding to the first countermeasure and the problem level set in the second state corresponding to the second countermeasure. conduct mediation with Also, for example, the countermeasure arbitration unit 40 arbitrates between the first countermeasure and the second countermeasure based on the presence or absence of production resources required for the first countermeasure or the second countermeasure. The presence or absence of production resources is managed by the resource database 41 .

The instruction output unit 50 outputs the first priority instruction. Specifically, the instruction output unit 50 outputs the first countermeasure or the second countermeasure, which is the first priority instruction, according to the arbitration by the countermeasure arbitration unit 40 . For example, until the effect determination unit 60, which will be described later, determines the effect of the executed first priority instruction, the instruction output unit 50 determines the second priority instruction, which is determined from among the plurality of countermeasures, and has the priority after the first priority instruction. Do not output priority instructions. Further, for example, when the first priority instruction is executed, the effect determination unit 60 changes the state of the production floor after execution of the first priority instruction to the state of the production floor before execution of the first priority instruction. If it is determined that there is an improvement tendency as described above, the instruction output unit 50 does not output the second priority instruction before execution, which is extracted in relation to the state on the production floor corresponding to the first priority instruction. Further, for example, when the first priority instruction is executed, the effect determination unit 60 changes the state of the production floor after execution of the first priority instruction to the state of the production floor before execution of the first priority instruction. If it is determined that there is no improvement tendency as described above, the countermeasure determination unit 30 determines the second priority instruction to be executed based on the priority of each of the plurality of countermeasures excluding the first priority instruction, and the instruction output unit 50 outputs the second priority indication. Further, for example, when the effect determination unit 60 cannot determine the effect for a predetermined period of time, the countermeasure determination unit 30 determines the second priority to be executed based on the priority of each of the plurality of countermeasures excluding the first priority instruction. After determining the instruction, the instruction output unit 50 outputs the second priority instruction. The instruction output unit 50 may output an instruction to a control unit of a production apparatus or a mobile terminal possessed by a worker, or through a production management apparatus that manages the production apparatus or a worker management apparatus that manages workers. Instructions may be output to production equipment and workers.

The effect determination unit 60 determines the effect of the executed countermeasure (instruction) based on the state of the production floor before and after the determined countermeasure (in other words, the output first priority instruction) was executed. Specifically, the effect determination unit 60 determines whether the first countermeasure or the second countermeasure corresponding to the output instruction is executed based on at least one of the first state and the second state before and after the first countermeasure or the second countermeasure is executed. Determine the effect of the countermeasure or the second countermeasure.

The update unit 70 updates the first and second conditions, that is, the learning models 23, 24, 33 and 34, based on the determined effects.

The details of the state monitoring unit 20, the countermeasure determination unit 30, the countermeasure mediation unit 40, the instruction output unit 50, the effect determination unit 60, and the update unit 70 will be described later.

Monitoring of the first state by the first state monitoring unit 21 and determination of the first countermeasure by the first countermeasure determination unit 31, monitoring of the second state by the second state monitoring unit 22 and second countermeasure by the second countermeasure determination unit 32 decisions are made in parallel. This will be described with reference to FIG.

FIG. 4 is a diagram schematically showing the flow of operations of the production floor management system 1 according to the embodiment.

The production floor management system 1 performs problem discovery by the state monitoring unit 20, determination of countermeasure policies by the countermeasure determination unit 30, arbitration by the countermeasure arbitration unit 40, and execution of countermeasures by the instruction output unit 50. These can be applied to so-called OODA loops, where problem finding corresponds to 'Observe', policy decision to 'Orient', mediation to 'Decide', and countermeasure execution to 'Action'. corresponds to

The monitoring of the first state by the first state monitoring unit 21 and the determination of the first countermeasure by the first countermeasure determining unit 31 are for improving the MTTR as described above. A cycle of problem discovery by the unit 21, determination of a countermeasure policy by the first countermeasure determination unit 31, arbitration by the countermeasure arbitration unit 40, and countermeasure execution by the instruction output unit 50 is defined as an MTTR cycle. The monitoring of the second state by the second state monitoring unit 22 and the determination of the second countermeasure by the second countermeasure determining unit 32 are for improving the MTBF as described above. A cycle of problem discovery by the unit 22, determination of a countermeasure policy by the second countermeasure determination unit 32, arbitration by the countermeasure arbitration unit 40, and countermeasure execution by the instruction output unit 50 is defined as an MTBF cycle.

In this way, the monitoring of the first state by the first state monitoring unit 21 and the determination of the first countermeasure by the first countermeasure determining unit 31, the monitoring of the second state by the second state monitoring unit 22 and the second countermeasure determining unit 32 The determination of the second countermeasure by is carried out in parallel. For example, process guarantees can be achieved by the OODA loop from both MTBF and MTTR.

In addition, each process of the OODA loop, problem finding, policy decision, arbitration, and countermeasure execution, maintains independence and is executed in parallel to minimize the waiting time between processes, enabling real-time control. can be realized. Details will be described later, but it is possible to implement countermeasures while changing the order of priority for the conditions on the production floor that change over time.

Although the first state is the production state of the production equipment and the second state is the production resource state, the present invention is not limited to this. For example, the first state may be the production state of the production device, and the second state may be the production state of the production device different from the first state. Also, for example, the first state may be the state of the production resource, and the second state may be the state of the production resource different from the first state. For example, although FIG. 4 shows an example in which the MTBF cycle and the MTTR cycle are performed in parallel, a plurality of MTBF cycles may be performed in parallel, or a plurality of MTTR cycles may be performed in parallel. may Note that when the MTBF cycle and the MTTR cycle are performed in parallel and the first and second countermeasures are arbitrated, the countermeasure arbitration unit 40 may prioritize the first countermeasure over the second countermeasure. In particular, when there is no difference in the degree of problem, which will be described below, the operation of the production equipment can be maintained by giving priority to the first countermeasure. The phrase "there is no difference in the degree of problem" as used herein includes that the degree of problem is the same and that the difference in the degree of problem is within a predetermined difference.

Next, the details of the state monitoring unit 20, the countermeasure determination unit 30, the countermeasure mediation unit 40, the instruction output unit 50, the effect determination unit 60, and the update unit 70 will be described with reference to FIGS. 5 to 12. FIG.

First, the details of the state monitoring unit 20 will be explained using FIG.

FIG. 5 is a flow chart showing an example of the operation of the state monitoring section 20 according to the embodiment. FIG. 5 shows problem finding (Observe) processing in the OODA loop.

First, the state monitoring unit 20 acquires monitoring data (step S11). Specifically, the first state monitoring unit 21 obtains monitoring data regarding the production state of the production equipment, and the second state monitoring unit 22 obtains monitoring data regarding the state of the production resource.

Next, the state monitoring unit 20 reads the learning model (step S12). Specifically, the first state monitoring unit 21 reads the learning model 23 and the second state monitoring unit 22 reads the learning model 24 .

The learning model 23 is a first learning model for detecting a first predetermined state that occurs on the production floor corresponding to the production state of the production equipment, and is associated with a detection threshold corresponding to the production state of the production equipment. For example, the learning model 23 is input with acquired monitoring data, and the acquired monitoring data indicates production error information occurring in a production apparatus, production volume information of a product produced by the production apparatus, and It is learned to output a first predetermined state (for example, productivity, quality, or scrap) corresponding to a production index related to at least one piece of quality information of a product produced by the production apparatus. Here, the learning method (update method) of the learning model 23 will be described with a specific example.

For example, it is assumed that a decrease in productivity is detected based on the detection threshold corresponding to the productivity associated with the learning model 23, and countermeasures for increasing productivity are output. As a result, productivity is increased and quality is decreased. For example, if the learning policy is set to emphasize quality, learning is performed so that the detection threshold corresponding to productivity is loosened (decrease in productivity is less likely to be detected) based on the above effect. . Thereby, a detection threshold that balances productivity and quality is learned.

The learning model 24 is a second learning model for detecting a second predetermined state occurring on the production floor corresponding to the state of the production resource, and is associated with a detection threshold corresponding to the state of the production resource. For example, the learning model 24 is input with acquired monitoring data, so that the operating index related to the operating state of the production equipment included in the production resource indicated by the acquired monitoring data, or the worker included in the production resource is learned to output a second predetermined state (eg, deterioration of a production device or facility element, or an error in an operator's work, etc.) corresponding to a work index related to the work performed by the . Here, the learning method (update method) of the learning model 24 will be described with a specific example.

For example, it is assumed that a decrease in the flow rate of the nozzle is detected based on the detection threshold corresponding to the flow rate of the nozzle associated with the learning model 24, and a countermeasure to perform maintenance of the nozzle within one week is output. . For example, it is assumed that the adsorption error rate worsens before maintenance is performed (before one week has passed) after countermeasures are output, and the output countermeasures are not effective. In this case, it is possible that the decrease in the flow rate of the nozzle was detected too late. learning is done in such a way that As a result, the detection threshold is learned so that maintenance can be performed at the optimum time. Note that the detection threshold corresponding to the flow rate of the nozzle is an example, and other detection thresholds may be set for the deviation of the tape stop position of the feeder, the deviation of the pickup position of the component sucked by the nozzle, or the deviation of the transfer position of the board. can also apply the above learning.

Next, the state monitoring unit 20 determines whether or not a problem has been detected in the first predetermined state or the second predetermined state (step S13). For example, the first state monitoring unit 21 determines whether or not it is detected that the productivity is declining, the quality is declining, or the number of defective products is increasing. For example, the second state monitoring unit 22 determines whether or not it has detected that the production equipment is deteriorating, that the equipment elements are deteriorating, or that there is an error in the worker's work.

If no problem is detected (No in step S13), the process from step S11 is repeated until a problem is detected.

If a problem is detected (Yes in step S13), a countermeasure policy for the detected problem is determined.

Next, the details of the countermeasure determination unit 30 will be explained using FIG.

FIG. 6 is a flow chart showing an example of the operation of the countermeasure determination unit 30 according to the embodiment. FIG. 6 shows the processing of policy determination (Orient) in the OODA loop.

The countermeasure determination unit 30 analyzes the production resources (for example, production resources that can be the cause) for the problem detected by the state monitoring unit 20 (step S21). For example, when the first state monitoring unit 21 detects a problem that the productivity is declining in the mounting process of the mounting apparatus, the production resources that can cause the problem are substrates, parts, workers, heads, nozzles, and so on. and feeders (see FIG. 3). Also, for example, when the second state monitoring unit 22 detects a problem of nozzle deterioration, it analyzes that the nozzle is the production resource that can cause the problem.

Next, the countermeasure determination unit 30 reads the learning model (step S22). Specifically, the first countermeasure determination unit 31 reads the learning model 33 and the second countermeasure determination unit 32 reads the learning model 34 .

The learning model 33 is a learning model for determining the first countermeasure corresponding to the production state of the production equipment. Specifically, learning for updating the priority for determining the priority of the first countermeasure. is a model. For example, the learning model 33 is trained to output countermeasures for the detected problem as candidates by inputting the detected problem. If multiple candidates are output for the detected problem, multiple countermeasures are extracted corresponding to the multiple candidates. A learning method of the learning model 33 will be described later.

The learning model 34 is a learning model for determining the second countermeasure corresponding to the state of production resources, and specifically, a learning model for updating the priority for determining the priority of the second countermeasure. is. For example, the learning model 34 is trained to output countermeasures for the detected problem as candidates by inputting the detected problem. If multiple candidates are output for the detected problem, multiple countermeasures are extracted corresponding to the multiple candidates. A learning method of the learning model 34 will be described later.

Next, the countermeasure determination unit 30 analyzes the priority for determining the priority of each of the multiple countermeasures (step S23). For example, when the learning models 33 and 34 output countermeasures for the detected problem as candidates, they are output in association with the priority of the countermeasures. Thereby, the countermeasure determination unit 30 can grasp the priority of each countermeasure.

Next, the countermeasure determination unit 30 creates a countermeasure candidate list (step S24). The countermeasure candidate list will be described with reference to FIG.

FIG. 7 is a table showing an example of a countermeasure candidate list according to the embodiment.

For example, it is assumed that a deterioration in the pick-up error rate of parts by nozzles is detected as a problem of the first predetermined state (for example, productivity, quality, or scrap). In this case, the learning model 33 outputs countermeasure candidates such as pickup position teaching, feeder replacement, and nozzle replacement, and also outputs a priority probability as the priority of each countermeasure candidate. As a result, the countermeasure determination unit 30 can create a countermeasure candidate list as shown in FIG. In the countermeasure candidate list shown in FIG. 7, the pick-up position teaching has a priority probability of 60% as a countermeasure candidate, and has the highest priority among the countermeasure candidates for the problem of deterioration of the pick-up error rate. Note that if there are countermeasure candidates with the same priority probability, it may be determined based on past results (countermeasure success count, transition of priority probability, etc.).

Returning to the description of FIG. 6, next, the countermeasure determination unit 30 registers the countermeasure candidate with the highest priority in the created countermeasure candidate list in the priority countermeasure list as a countermeasure against the detected problem (step S25). For example, when the countermeasure candidate list shown in FIG. 7 is created, the countermeasure of suction position teaching is registered in the priority countermeasure list for the problem of deterioration of the suction error rate. For example, a plurality of countermeasures extracted corresponding to the state on the production floor when the problem of deterioration of the pick-up error rate is detected are pick-up position teaching, feeder replacement, and nozzle replacement. When the countermeasure determination unit 30 registers the countermeasure of suction position teaching in the priority countermeasure list, it is a first priority instruction to be executed from among the plurality of countermeasures based on the order of priority (priority) of each of the plurality of countermeasures. , means that the adsorption position teaching is determined.

It should be noted that acquisition of monitoring data is repeated at predetermined time intervals, and various problems can be detected. For example, one problem may be detected before the remedy for another problem is completed. For example, multiple problems with a first predetermined condition (e.g., productivity, quality, or work errors, etc.) may be detected. In this way, each time a problem is detected, a countermeasure candidate list for that problem is created, and the countermeasure with the highest priority for each countermeasure candidate list is added to the priority countermeasure list. FIG. 8 shows an example of the priority countermeasure list.

FIG. 8 is a table showing an example of a priority countermeasure list according to the embodiment.

As shown in FIG. 8, it can be seen that, as a problem for the first predetermined state, a countermeasure of suction position teaching is registered in the priority countermeasure list for the problem of deterioration of the suction error rate described above. As for the problem of the second predetermined state, it can be seen that the countermeasure of cleaning is registered in the priority countermeasure list for the problem of feeder sliding failure. As for the problem of the second predetermined state, it can be seen that the countermeasure of replacing the conveyor belt is registered in the priority countermeasure list for the problem of wear of the conveyor belt. Further, a problem level is set in advance as an index indicating the seriousness of the problem for the first predetermined state and the second predetermined state. For example, the problem of deterioration of the suction error rate can be said to be the problem of the first state (the production state of the production apparatus) when the problem is detected, so the problem level is set to the problem of deterioration of the suction error rate. can be said to be the degree of problem set in the first state when the problem is detected. Further, for example, the problem of feeder sliding failure can be said to be a problem in the second state (the state of the production resource) when the problem is detected, so the problem level is set to the problem of feeder sliding failure. can be said to be the degree of problem set in the second state when the problem is detected.

Next, the details of the countermeasure arbitration unit 40 will be explained using FIG.

FIG. 9 is a flow chart showing an example of the operation of the countermeasure arbitration unit 40 according to the embodiment. FIG. 9 shows arbitration (Decide) processing in the OODA loop.

First, the countermeasure arbitration unit 40 reads the prioritized countermeasure list (step S31) and selects countermeasures with high priority (that is, degree of problem) (step S32). For example, when the read priority measure list is the list shown in FIG. 8, the measure arbitration unit 40 selects suction position teaching as a measure with a high priority.

Next, the countermeasure arbitration unit 40 reads resource data (resource management table) from the resource database 41 (step S33). The resource management table will be explained using FIG.

FIG. 10 is a table showing an example of a resource management table according to the embodiment.

For example, in the resource management table, the presence or absence of each production resource, specifically, whether each production resource is currently taking some measures, or whether each production resource is currently taking any measures It is managed whether or not In FIG. 10, the presence or absence of production resources is indicated by locking on and off. The resource management table shown in FIG. 10 manages the presence or absence of heads, nozzles, feeders, and workers A and B as production resources. The head and feeder are currently under some kind of countermeasure and the lock is on. The nozzle is currently unlocked as no countermeasures are being taken. Workers A and B are currently unlocked because they are not taking any countermeasures. The lock referred to here may include either real space or virtual space, or both. For example, prohibiting removal from a production device until the lock is turned off for a feeder whose resources are locked means locking in the real space. Also, prohibiting the use of a feeder in which a resource is locked when creating or updating a production plan means locking in virtual space.

Returning to the description of FIG. 9, next, the countermeasure arbitration unit 40 determines whether or not the production resource is locked for the selected countermeasure (step S34). For example, it is assumed that the countermeasure mediation unit 40 selects suction position teaching. Also, for example, it is assumed that the suction position teaching is a countermeasure that the worker A takes. In this case, the countermeasure arbitration unit 40 confirms the lock state of worker A in the read resource management table.

If the production resource for the selected countermeasure is locked (Yes in step S34), the countermeasure arbitration unit 40 selects the countermeasure with the next highest priority (step S35), and performs the processing from step S33 again.

If the production resource for the selected countermeasure is not locked (No in step S34), the countermeasure arbitration unit 40 locks the production resource for the selected countermeasure (step S36).

Then, it is determined whether or not the production resources for all the measures included in the priority measure list are locked (step S37). If the production resources for all countermeasures are not locked (No in step S37), the process from step S32 is performed again except for the countermeasure selected this time. If the production resources for all the measures are locked (Yes in step S37), the selected high-priority measures among the measures in the prioritized measure list whose production resources were not locked are executed. will be

Next, the details of the instruction output unit 50, the effect determination unit 60, and the update unit 70 will be described using FIG.

FIG. 11 is a flow chart showing an example of operations of the instruction output unit 50, the effect determination unit 60, and the updating unit 70 according to the embodiment. FIG. 11 shows countermeasure execution (Action) in the OODA loop and processing after the countermeasure is executed. In the following, the case where the countermeasure selected by the countermeasure arbitration unit 40 for the problem of the worsening of the suction error rate is suction position teaching, and the locked production resource is the worker A will be described as the specific example 1. A specific example 2 is a case where the countermeasure selected by the countermeasure arbitration unit 40 for the feeder sliding failure problem is cleaning, and the locked production resource is the feeder and the worker B. FIG.

First, the instruction output unit 50 executes a countermeasure for the production resource locked by the countermeasure arbitration unit 40 (step S41). For example, in the case of Specific Example 1, the instruction output unit 50 outputs a first countermeasure (for example, a first priority instruction) that causes the worker A to perform the pickup position teaching. For example, in the case of Specific Example 2, the instruction output unit 50 outputs a second countermeasure (for example, a first priority instruction) to have worker B clean the feeder. In this way, instructions corresponding to countermeasures are output and executed. Execution of instructions may be performed manually by an operator or the like, or may be performed automatically by a production device or the like.

Next, the effect determination unit 60 acquires monitoring data (step S42). Specifically, the effect determination unit 60 acquires monitoring data regarding the production status of production equipment or monitoring data regarding the status of production resources. The reason why the effect determination unit 60 acquires the monitoring data is to confirm the change in the state on the production floor due to the execution of the instruction according to the countermeasure, that is, to determine the effect of the instruction according to the executed countermeasure. is. For example, in the case of Specific Example 1, the effect determination unit 60 acquires monitoring data about the result of the mounting process regarding suction. That is, the effect determination unit 60 monitors the tendency of the error information related to the monitored nozzle as monitoring data. For example, in the case of specific example 2, the effect determination unit 60 acquires monitoring data about the state of the feeder. That is, the effect determination unit 60 monitors the tendency of the error information related to the monitored feeder as monitoring data. For example, the effect determination section 60 detects the first predetermined state or the second predetermined state using the learning model 23 or 24 in the same way as the state monitoring section 20 does.

Next, the effect determination unit 60 determines whether or not a problem has been detected in the first predetermined state or the second predetermined state (step S43). When a problem is detected, it can be determined that there is no effect of the instructions according to the measures taken, or that the effects of the instructions according to the measures taken have not yet appeared, and the problem is detected. If not, it can be determined that the instruction was effective in accordance with the countermeasures taken.

If no problem is detected (No in step S43), the updating unit 70 updates the learning model 33 or 34 based on the determined effect (step S44). For example, in the case of Concrete Example 1, when the problem is no longer detected, the updating unit 70 determines that the suction position teaching was effective against the problem, and the priority of the suction position teaching is increased. The learning model 33 is updated as follows. For example, in the case of Specific Example 2, when the problem is no longer detected, the updating unit 70 determines that cleaning was effective against the problem, and sets the learning model so that cleaning has a higher priority. Update 34.

Next, since the instruction of this time is completed, the effect determination unit 60 unlocks the production resource that was locked when executing the instruction of this time (step S45). For example, in the case of the specific example 1, the locked worker A is unlocked. For example, in the case of the specific example 2, the locked operator B and the feeder are unlocked.

Next, the effect determination unit 60 deletes the measures that have been executed according to this instruction from the prioritized measures list (step S46). For example, in the case of specific example 1, the pickup position teaching is deleted from the priority countermeasure list. For example, in the case of specific example 2, cleaning is deleted from the priority measure list.

Then, the effect determination unit 60 deletes the countermeasure candidate list for the problem that is no longer detected by this instruction (step S47). For example, in the case of Specific Example 1, the problem of deterioration of the suction error rate is no longer detected, and countermeasures for this problem are unnecessary, so the countermeasure candidate list shown in FIG. 7 is deleted. Deletion of the countermeasure candidate list means that the effect determination unit 60 determines that the first priority instruction (instruction to teach the pickup position) is executed, and the state (suction error rate) before execution of the instruction to teach the pickup position. When it is determined that the suction error rate after execution of the suction position teaching instruction is improved by a predetermined amount or more, the instruction output unit 50 outputs the extracted suction error rate corresponding to the suction position teaching instruction. , means that the second priority instruction (instruction of feeder replacement or nozzle replacement) before execution is not output. In addition, the effect determination unit 60 may also determine whether a problem presented in the prioritized countermeasure list continues in addition to the problem for which countermeasures have been taken. The effect determination unit 60 may delete the prioritized countermeasure list for problems for which no problems are detected other than the problems for which countermeasures have been taken. Some of the problems detected are relevant and efficient countermeasures can be taken against such problems.

On the other hand, if a problem is detected (Yes in step S43), the effect determination unit 60 determines whether or not timeout has occurred (step S48). In other words, the effect determination unit 60 determines whether or not the effect cannot be determined for a predetermined period of time. Since it may take some time for the effect to appear after the instruction is executed, the process in step S48 is performed. For the predetermined period, for example, the time required for harvesting the effect is set for each instruction. In addition, such as measures added to the production plan and maintenance plan, by creating a plan that does not take immediate action, the production resource is unlocked and the effect judgment is executed. may be done after

If the timeout has not occurred (No in step S48), the processes in steps S42, S43, and S48 are repeated until the problem is no longer detected or the timeout occurs.

When timed out (Yes in step S48), the update unit 70 updates the learning model 33 or 34 based on the determined effect (step S49). For example, in the case of Concrete Example 1, if a timeout occurs while a problem is being detected, the updating unit 70 determines that the suction position teaching was not effective against the problem, and the priority of the suction position teaching is set to Update the learning model 33 so that it becomes lower. For example, in the case of Concrete Example 2, if a timeout occurs while a problem is detected, the updating unit 70 determines that cleaning was not effective against the problem, and lowers the priority of cleaning. Update the learning model 34. In addition, when a problem is detected but there is an improvement in the trend of the monitoring data (there is an improvement trend below a predetermined level), the updating unit 70 updates the learning model 34 so that the degree of priority decrease is small. to update.

Next, since the current instruction has been completed, the effect determination unit 60 unlocks the production resources that were locked when executing the current instruction (step S50). For example, in the case of the specific example 1, the locked worker A is unlocked. For example, in the case of the specific example 2, the locked operator B and the feeder are unlocked.

Next, the effect determination unit 60 deletes the countermeasures that have been executed according to this instruction from the priority countermeasure list (step S51). For example, in the case of specific example 1, the pickup position teaching is deleted from the priority countermeasure list. For example, in the case of specific example 2, cleaning is deleted from the priority measure list.

Next, the effect determination unit 60 updates the countermeasure candidate list for problems that remain detected even after the current instruction is executed (step S52). Here, updating of the countermeasure candidate list in the case of specific example 1 will be described with reference to FIG. 12 .

FIG. 12 is a table showing an example of the updated countermeasure candidate list according to the embodiment.

For example, if the problem of deterioration of the suction error rate is still detected even if the instruction of the suction position teaching is executed, as shown in FIG. 12, the suction position teaching is deleted from the list of candidate measures. In addition, some time has passed since the countermeasure candidate list shown in FIG. 7 was created, and during that time production continues on the production floor. The priority of nozzle replacement may also have changed. Therefore, the effect determination unit 60 may analyze the priority and update the priority in the countermeasure candidate list. For example, in the countermeasure candidate list shown in FIG. 7, the ratio of priority (priority probability) between feeder replacement and nozzle exchange is 3:1, but in the countermeasure candidate list shown in FIG. I know it's changing.

Returning to the description of FIG. 11, the effect determination unit 60 registers the highest-priority measure candidate in the created measure candidate list in the priority measure list as a measure against the detected problem (step S53). For example, when the countermeasure candidate list shown in FIG. 12 is created, a countermeasure of exchanging nozzles is registered in the priority countermeasure list for the problem of deterioration of the suction error rate. In other words, an instruction to replace the nozzle can be output in response to the problem that the suction error rate continues to deteriorate even after the suction position teaching is performed.

This is because the operations after step S50 are not executed until the effect determination unit 60 determines the effect of the executed first priority instruction (suction position teaching). , feeder exchange and nozzle exchange), and the second priority instruction (feeder exchange or nozzle exchange) after the pickup position teach instruction is not output. This is because the instruction to replace the nozzle is output after it is determined that the suction position teaching is ineffective.

In addition, the effect determination unit 60 determines that the state (suction error rate) before execution of the suction position teaching is compared with the state (suction error rate) after the execution of the suction position teaching due to the execution of the first priority instruction (suction position teaching). When it is determined that the suction error rate does not tend to improve beyond a predetermined level, the countermeasure determination unit 30 determines the first countermeasure to be executed based on the priority of each of the plurality of countermeasures (feeder replacement and nozzle replacement) excluding the suction position teaching. This means that the 2nd priority instruction (nozzle replacement) is determined, and the instruction output unit 50 outputs the nozzle replacement instruction. This is because instructions to replace nozzles are output with priorities after the suction position teaching, which has the highest priority.

Further, when the effect determination unit 60 cannot determine the effect for a predetermined period of time, the countermeasure determination unit 30 selects a plurality of countermeasures (feeder exchange instruction and nozzle exchange instruction) excluding the first priority instruction (suction position teaching). ), the second priority instruction (nozzle replacement) to be executed is determined, and the instruction output unit 50 outputs the nozzle replacement instruction. This is because, after the time-out, an instruction to replace the nozzle with the highest priority among the plurality of countermeasures excluding the suction position teaching is output.

In step S13, after a problem is detected by the status monitoring unit 20 and a high-priority countermeasure is registered in the priority countermeasure list from the countermeasure candidate list for the problem, the registered countermeasure may have a low priority. In some cases, the registered countermeasures may not be executed easily because the production resources for the registered countermeasures have already been locked. In such a case, the previously executed countermeasures for other problems may solve the problems corresponding to the unexecuted countermeasures. In this case, even before the registered countermeasure is executed, it may be deleted from the priority countermeasure list and the countermeasure candidate list may also be deleted.

As described above, the first and second countermeasures are determined to deal with the different first and second conditions on the production floor, and the first and second countermeasures are arbitrated. By doing so, it is possible to output the optimum first or second countermeasure. In this way, according to the production floor management system 1 of the present disclosure, it is possible to output the optimum work instructions from a plurality of coping methods.

(Other embodiments)
Although the production floor management system 1 of the present disclosure has been described above based on the embodiments, the present disclosure is not limited to the above embodiments. As long as it does not deviate from the spirit of the present disclosure, modifications that can be made by those skilled in the art to the present embodiment, and forms constructed by combining the components of different embodiments are also included within the scope of the present disclosure. be

For example, in the above embodiment, the production floor management system 1 includes the effect determination unit 60, the update unit 70, and the learning models 23, 34, 33, and 34, but it does not have to.

For example, in the above embodiment, the first state is the production state of the production equipment and the second state is the production resource state, but the present invention is not limited to this. For example, the first state and the second state are not particularly limited as long as they are different states on the production floor.

For example, in the above embodiment, the production floor management system 1 includes the acquisition unit 10, the state monitoring unit 20, the countermeasure determination unit 30, the countermeasure mediation unit 40, the instruction output unit 50, the effect determination unit 60, and the update unit 70. , some of which may be included in the production equipment. For example, a production device may include the acquisition unit 10 , the state monitoring unit 20 , and the countermeasure determination unit 30 .

For example, instructions according to countermeasures may consist of multiple instructions. In addition, instructions corresponding to countermeasures may be output to a plurality of production apparatuses or mobile terminals possessed by a plurality of workers.

For example, the present disclosure can be realized not only as the production floor management system 1, but also as a work instruction method including steps (processes) performed by each component constituting the production floor management system 1.

FIG. 13 is a flow chart showing an example of a work instruction method according to another embodiment.

The work instruction method is a work instruction method in a production floor management system that manages the state of a production floor equipped with production equipment that produces products, and as shown in FIG. 13, the first state of the production floor is monitored. (Step S1), a second state different from the first state on the production floor is monitored (Step S2), a first countermeasure corresponding to the first state is determined (Step S3), and a second countermeasure corresponding to the second state is determined (Step S3). Two countermeasures are determined (step S4), the first countermeasure and the second countermeasure are arbitrated (step S5), and an instruction corresponding to the first countermeasure or the second countermeasure is output according to the arbitration (step S6). include.

For example, the steps in the work instruction method may be executed by a computer (computer system). Further, the present disclosure can be realized as a program for causing a computer to execute the steps included in the work instruction method. Furthermore, the present disclosure can be implemented as a non-temporary computer-readable recording medium such as a CD-ROM recording the program.

For example, when the present disclosure is implemented by a program (software), each step is executed by executing the program using hardware resources such as the CPU, memory, and input/output circuits of the computer. . That is, each step is executed by the CPU acquiring data from a memory, an input/output circuit, or the like, performing an operation, or outputting the operation result to the memory, an input/output circuit, or the like.

Also, each component included in the production floor management system 1 of the above embodiment may be realized as a dedicated or general-purpose circuit.

Also, each component included in the production floor management system 1 of the above embodiment may be implemented as an LSI (Large Scale Integration), which is an integrated circuit (IC).

Also, the integrated circuit is not limited to an LSI, and may be realized by a dedicated circuit or a general-purpose processor. A programmable FPGA (Field Programmable Gate Array) or a reconfigurable processor capable of reconfiguring connections and settings of circuit cells inside the LSI may be used.

Furthermore, if a technology for integrating circuits to replace LSIs emerges due to advances in semiconductor technology or another technology derived from it, then naturally each component included in the production floor management system 1 will be integrated into circuits using that technology. may be broken.

In addition, forms obtained by applying various modifications to the embodiments that a person skilled in the art can think of, and forms realized by arbitrarily combining the components and functions in each embodiment within the scope of the present disclosure are also included in this disclosure.

The present disclosure can be used, for example, for managing production floors.

1 production floor management system 2 communication network 4 mounting line 5 management device 10 acquisition unit 20 state monitoring unit 21 first state monitoring unit 22 second state monitoring unit 23, 24, 33, 34 learning model 30 countermeasure determination unit 31 first countermeasure Determination unit 32 Second countermeasure determination unit 40 Countermeasure mediation unit 41 Resource database 50 Instruction output unit 60 Effect determination unit 70 Update unit M1 Board supply device M2 Board delivery device M3 Printer M4, M5 Mounting device M6 Reflow device M7 Board recovery device

Claims (14)

1. A production floor management system for managing conditions on a production floor comprising production equipment that produces products, comprising:
   a first condition monitor for monitoring a first condition on the production floor;
   a first countermeasure determination unit that determines a first countermeasure corresponding to the first state;
   a second condition monitoring unit that monitors a second condition different from the first condition on the production floor;
   a second countermeasure determination unit that determines a second countermeasure corresponding to the second state;
   a countermeasure arbitration unit that arbitrates between the first countermeasure and the second countermeasure;
   an instruction output unit that outputs an instruction corresponding to the first countermeasure or the second countermeasure according to the arbitration.
2. Monitoring of the first state by the first state monitoring section and determination of the first countermeasure by the first countermeasure determining section; Monitoring of the second state by the second state monitoring section and by the second countermeasure determining section The production floor management system according to claim 1, wherein the determination of the second countermeasure is performed in parallel.
3. The countermeasure arbitration unit performs the arbitration based on the degree of problem set in the first state corresponding to the first countermeasure and the degree of problem set in the second state corresponding to the second countermeasure. 3. A production floor management system according to Item 1 or 2.
4. If there is no difference between the problem level set in the first state corresponding to the first countermeasure and the problem level set in the second state corresponding to the second countermeasure, the countermeasure mediation unit 4. The production floor management system according to claim 3, wherein said arbitration is performed with priority given to countermeasures.
5. The production floor management system performs the executed second countermeasure based on at least one of the first state and the second state before and after the first countermeasure or the second countermeasure corresponding to the output instruction is executed. 5. The production floor management system according to any one of claims 1 to 4, further comprising an effect determination unit that determines the effect of the first countermeasure or the second countermeasure.
6. the first state is a production state of the production apparatus;
   The production floor management system according to any one of claims 1 to 5, wherein the second status is the status of production resources managed on the production floor and used for production.
7. 7. The production floor management system according to claim 6, wherein said countermeasure arbitration unit performs said arbitration based on the presence or absence of said production resource required for said first countermeasure or said second countermeasure.
8. The production floor management system further comprises a first learning model for detecting a first predetermined condition occurring on the production floor corresponding to the production condition;
   8. The production floor management system according to claim 6, wherein said first condition monitoring unit detects said first predetermined condition based on said first learning model.
9. The first state monitoring unit provides, within a predetermined period of time, production error information occurring in the production apparatus, production volume information of the product produced by the production apparatus, and quality information of the product produced by the production apparatus. detecting the first predetermined condition corresponding to at least one production index;
   9. The production floor management system according to claim 8, wherein said first countermeasure determination unit determines said first countermeasure including a countermeasure for improving said production index.
10. The production floor management system further comprises a second learning model for detecting a second predetermined condition occurring on the production floor corresponding to the condition of the production resource;
    The production floor management system according to any one of claims 6 to 9, wherein the second state monitoring unit detects the second predetermined state based on the second learning model.
11. The second state monitoring unit detects the second predetermined state corresponding to an operation index relating to the operation state of the production equipment included in the production resource,
    11. The production floor management system according to claim 10, wherein said second countermeasure determination unit determines said second countermeasure including countermeasures for said production equipment corresponding to said operation index.
12. The second state monitoring unit detects the second predetermined state corresponding to a work index related to a work performed by a worker included in the production resource,
    12. The production floor management system according to claim 10, wherein said second countermeasure determination unit determines said second countermeasure including work of said worker corresponding to said work index.
13. A work instruction method in a production floor management system that manages the state of a production floor equipped with production equipment that produces products,
    monitor a first condition on the production floor;
    monitor a second condition on the production floor that is different from the first condition;
    determining a first countermeasure corresponding to the first state;
    determining a second countermeasure corresponding to the second state;
    mediating between the first measure and the second measure;
    A work instruction method, including outputting an instruction corresponding to the first countermeasure or the second countermeasure according to the arbitration.
14. A work instruction program that causes a computer to execute the work instruction method according to claim 13.

Images