Production assistance device, production assistance method, and program

Abstract

A production assistance device (100) comprises: an acquisition unit (60) that acquires a plurality of types of mounting logs relating to mounting machinery (200); an analysis unit (100a) that has a plurality of analysis engines (for example, a first estimation processing unit (10) to a fourth estimation processing unit (40)) that analyze each of the plurality of types of mounting logs, each of the plurality of analysis engines analyzing a plurality of malfunction factors relating to the mounting machinery (200); and an assessment unit (for example, a comprehensive assessment unit (50)) that, on the basis of the analysis results from each of the plurality of analysis engines, calculates the possibility that each of the plurality of malfunction factors is a factor in malfunction of the mounting machinery (200).

Description

Production support device, production support method and program

[0001] The present disclosure relates to a production support device, a production support method, and a program.

[0002] Patent Document 1 discloses a production support method that can present the degree of impact that an alert generated in a unit of production equipment has on production in a production system.

[0003] JP 2023-82616 A

[0004] When a mounting machine malfunctions, it is desirable to be able to more accurately estimate the cause.

[0005] Therefore, the present disclosure provides a production support device, a production support method, and a program that can more accurately estimate the cause of a malfunction in a mounting machine.

[0006] A production support device according to one aspect of the present disclosure comprises an acquisition unit that acquires multiple types of mounting logs related to a mounting machine, an analysis unit having multiple analysis engines that analyze each of the multiple types of mounting logs, each of the multiple analysis engines analyzing multiple malfunction factors related to the mounting machine, and a determination unit that calculates the likelihood that each of the multiple malfunction factors is the cause of the malfunction of the mounting machine based on the analysis results of each of the multiple analysis engines.

[0007] A production support method according to one aspect of the present disclosure includes a plurality of analysis engines that acquire multiple types of mounting logs related to a mounting machine and analyze each of the multiple types of mounting logs, each of which analyzes multiple malfunction factors related to the mounting machine, and calculates the likelihood that each of the multiple malfunction factors is the cause of the malfunction of the mounting machine based on the results of the analysis by the multiple analysis engines.

[0008] A program according to one aspect of the present disclosure is a program for causing a computer to execute the above-described production support method.

[0009] According to one aspect of the present disclosure, it is possible to realize a production support device or the like that can more accurately estimate the cause of a malfunction in a mounting machine.

[0010] FIG. 1 is a diagram showing the configuration of a production support system according to an embodiment. FIG. 2 is a diagram showing the configuration of a first estimation processing unit according to an embodiment. FIG. 3 is a diagram showing the configuration of a second estimation processing unit according to an embodiment. FIG. 4 is a diagram showing the configuration of a third estimation processing unit according to an embodiment. FIG. 5 is a diagram showing the configuration of a fourth estimation processing unit according to an embodiment. FIG. 6 is a flowchart showing the operation of a production support device according to an embodiment. FIG. 7 is a diagram showing an example of display of a comprehensive judgment result of a production support device according to an embodiment.

[0011] (Background to the present disclosure) Conventionally, in order to identify the cause of a malfunction of production equipment such as a mounting machine, production logs such as mounting logs have been analyzed. When one-sided production logs among multiple types of production logs are analyzed using a single analysis engine, the analytical perspective may become biased, resulting in insufficient analysis. Furthermore, there are problems in that it is not possible to confirm the correlation between multiple malfunction estimation results, making it difficult to prioritize multiple malfunction candidate locations.

[0012] Therefore, the inventors of the present application have conducted extensive research into production support devices that can more accurately estimate the causes of malfunctions in mounting machines, and have devised the production support devices described below. Furthermore, the inventors of the present application have devised a production support device that can assign priorities to multiple potential malfunction locations.

[0013] A malfunction refers to the occurrence of an alert that does not necessarily result in the stopping of the mounting machine, and is, for example, a state in which it is suspected that at least one of the feeder, head spindle, and nozzle has failed or deteriorated.

[0014] A production support device according to a first aspect of the present disclosure comprises an acquisition unit that acquires multiple types of mounting logs related to a mounting machine, an analysis unit having multiple analysis engines that analyze each of the multiple types of mounting logs, each of the multiple analysis engines analyzing multiple malfunction factors related to the mounting machine, and a determination unit that calculates the likelihood that each of the multiple malfunction factors is the cause of the malfunction of the mounting machine based on the analysis results of each of the multiple analysis engines.

[0015] This allows multiple types of mounting logs to be analyzed using multiple analysis engines, and the analysis results to be combined to calculate the cause of the malfunction, preventing bias in the analytical perspective.As a result, the production support device can more accurately estimate the cause of the malfunction of the mounting machine.

[0016] Also, for example, the production support device according to the second aspect may be the production support device according to the first aspect, and the multiple types of implementation logs may include at least two of environmental data, event logs, defect rates, quality data, and sensor data.

[0017] This allows the cause of the malfunction of the mounting machine to be more accurately estimated using the mounting log that is normally acquired by the mounting machine.

[0018] Also, for example, the production support device according to the third aspect may be the production support device according to the first or second aspect, and the multiple malfunction factors may include at least two of equipment factors, work factors, part factors, and environmental factors.

[0019] This allows accurate estimation of the likelihood of at least two of the equipment factors, work factors, part factors, and environmental factors.

[0020] Furthermore, for example, the production support device according to the fourth aspect may be a production support device according to any one of the first to third aspects, and the determination unit may cause the display unit to display the calculated possibility of the cause of the malfunction of the mounting machine for each of the plurality of malfunction factors.

[0021] This allows the display unit to display the possible causes of the malfunction, thereby helping the manager of the production support device or mounting machine to deal with the malfunction of the mounting machine.

[0022] Also, for example, a production support device according to a fifth aspect may be the production support device according to the fourth aspect, wherein the determination unit causes the display unit to display the analysis results by each of the plurality of analysis engines as a table.

[0023] This allows the manager of the production support device or mounting machine, etc., to compare the estimated results by checking the table, making it easy to understand the validity of the cause of the malfunction, etc. This effectively supports the manager, etc., in making judgments about the malfunction of the mounting machine.

[0024] Also, for example, a production support device according to a sixth aspect may be a production support device according to any one of the first to fifth aspects, and the plurality of analysis engines may be configured to analyze the plurality of types of implementation logs in parallel.

[0025] This allows multiple types of mounting logs to be analyzed in parallel, improving the usability of the production support device.

[0026] Furthermore, for example, a production support device according to a seventh aspect may be a production support device according to any one of the first to sixth aspects, and each of the plurality of analysis engines may be configured to analyze different types of implementation logs.

[0027] This makes it possible to further prevent the analytical viewpoint from becoming biased. Therefore, the production support device can more accurately estimate the cause of the malfunction of the mounting machine.

[0028] Furthermore, for example, a production support device according to an eighth aspect may be a production support device according to any one of the first to seventh aspects, wherein the plurality of analysis engines include a first analysis engine and a second analysis engine, the plurality of types of implementation logs include a first type log and a second type log that are different from each other, the first analysis engine takes the first type log as input and analyzes each of two or more first malfunction factors that correspond to the first type log among the plurality of malfunction factors, and the second analysis engine takes the second type log as input and analyzes each of two or more second malfunction factors that correspond to the second type log among the plurality of malfunction factors and that differ at least in part from the two or more first malfunction factors.

[0029] This allows for even a factor that cannot be estimated by analysis using one type of mounting log to be estimated by analysis using other mounting logs, thereby enabling a more accurate estimation of the cause of the malfunction of the mounting machine.

[0030] Furthermore, for example, a production support device according to a ninth aspect may be the production support device according to the eighth aspect, wherein the plurality of malfunction factors include a first malfunction factor and a second malfunction factor, the first malfunction factor is a factor corresponding to each of the first type of log and the second type of log, and the second malfunction factor is a factor corresponding to the first type of log, and the determination unit calculates the possibility of the first malfunction factor based on the possibility of the first malfunction factor calculated by the first analysis engine and the possibility of the first malfunction factor calculated by the second analysis engine, and calculates the possibility of the second malfunction factor based on the possibility of the first malfunction factor calculated by the first analysis engine.

[0031] As a result, information that may become noise (here, the second type of log) is not used in the analysis of the second malfunction cause, so the cause of the malfunction of the mounting machine can be estimated more accurately.

[0032] Also, for example, a production support device according to a tenth aspect may be the production support device according to the ninth aspect, wherein the determination unit calculates the possibility of the second malfunction factor based on the possibility of the first malfunction factor calculated by the first analysis engine and the possibility corresponding to the second analysis engine that has been set in advance.

[0033] By appropriately setting the preset value, it may be possible to more accurately estimate the cause of the malfunction of the mounting machine.

[0034] A production support method according to an eleventh aspect of the present disclosure includes acquiring multiple types of mounting logs related to a mounting machine, and having multiple analysis engines that analyze each of the multiple types of mounting logs, each of which analyzes multiple malfunction factors related to the mounting machine, and calculating the likelihood that each of the multiple malfunction factors is the cause of the malfunction of the mounting machine based on the results of the analysis by the multiple analysis engines.

[0035] This provides the same effects as the above-mentioned production support device.

[0036] A program according to a twelfth aspect of the present disclosure is a program for causing a computer to execute the above-described production support method.

[0037] This provides the same effects as the above-mentioned production support device.

[0038] These general or specific aspects may be realized as a system, a method, an integrated circuit, a computer program, or a non-transitory recording medium such as a computer-readable CD-ROM, or as any combination of the system, method, integrated circuit, computer program, or recording medium. The program may be pre-stored in the recording medium, or may be supplied to the recording medium via a wide area communication network including the Internet.

[0039] Hereinafter, the embodiments will be specifically described with reference to the drawings.

[0040] The embodiments described below are all comprehensive or specific examples. The numerical values, shapes, components, component placement and connection configurations, steps, and step order shown in the following embodiments are merely examples and are not intended to limit the present disclosure. Furthermore, among the components in the following embodiments, components not described in independent claims are described as optional components.

[0041] Furthermore, each figure is a schematic diagram and is not necessarily an exact illustration. Therefore, for example, the scales of the figures do not necessarily match. Furthermore, in each figure, substantially the same components are given the same reference numerals, and redundant explanations are omitted or simplified.

[0042] Furthermore, in this specification, terms indicating relationships between elements such as "same," as well as numerical values ​​and numerical ranges, are not expressions that only express a strict meaning, but are expressions that also include a substantially equivalent range, for example, a difference of about several percent (or about 10%).

[0043] Furthermore, in this specification, ordinal numbers such as "first" and "second" do not refer to the number or order of components unless otherwise specified, but are used for the purpose of avoiding confusion and distinguishing between components of the same type.

[0044] (Embodiment) Hereinafter, a production support device according to the present embodiment will be described with reference to FIGS.

[0045] [1. Configuration of the Production Support System] First, the configuration of a production support system including a production support device according to this embodiment will be described with reference to Figures 1 to 5. Figure 1 is a diagram showing the configuration of a production support system 1 according to this embodiment.

[0046] 1 , the production support system 1 includes a production support device 100, a mounting machine 200, and a display device 300. The production support device 100 and the mounting machine 200 are communicatively connected, and the production support device 100 and the display device 300 are communicatively connected. Note that, below, an example will be described in which the production support system 1 includes the mounting machine 200, which is an example of a production device, but the production support system 1 may also be used for production devices other than the mounting machine 200.

[0047] The production support device 100 is an information processing device that calculates the likelihood of a cause of a malfunction of the mounting device 200 based on multiple types of mounting logs related to the mounting device 200. As will be described in detail below, the production support device 100 analyzes the multiple types of mounting logs using multiple analysis engines, each of which has the function of estimating the cause of the malfunction, and estimates the cause of the malfunction by comprehensively determining the analysis results of each of the multiple analysis engines. Specifically, the production support device 100 has multiple analysis engines for analyzing the multiple types of mounting logs, each of which receives at least some different types of mounting logs, and each of the multiple analysis engines analyzes multiple malfunction factors related to the malfunction of the mounting device 200, and calculates the likelihood of a cause of the malfunction of the mounting device 200 by comprehensively determining the analysis results.

[0048] In this way, by using multiple analysis engines, it is possible to further optimize the analysis engines depending on the type of log input, thereby improving the accuracy of estimating malfunctions.

[0049] First, the mounting machine 200 and the display device 300 will be described, and then the production support device 100 will be described with reference to FIGS. 2 to 5 in addition to FIG.

[0050] The mounting machine 200 is an example of production equipment that makes up a production line. A production line (e.g., a mounting line) is configured to include one or more mounting machines 200. The mounting machine 200 is mounting equipment (component mounting device) that mounts components on objects such as boards. The mounting machine 200 is configured from multiple units, each having one function in component mounting, and each of the multiple units is configured from multiple parts to achieve that function.

[0051] In this embodiment, the mounting machine 200 includes units such as a drive control unit, a component mounting head unit, and a component supply unit. The drive control unit controls the rotation and movement of the component mounting head unit. The drive control unit includes a head spindle, which controls the rotation of the component mounting head unit by driving a motor. The component mounting head unit mounts (mounts) components supplied to the component supply position (component suction position) of the component supply unit onto a board. The component mounting head unit includes a mounting head (head) equipped with component suction nozzles (nozzles) that can pick up components from the component supply unit and move up and down individually. The head may be equipped with, for example, multiple nozzles. The component supply unit includes one or more feeders arranged side by side, each of which supplies components to the component supply position. In this embodiment, the feeders are tape feeders, but they may also be bulk feeders or the like.

[0052] The components are electronic components, such as, but not limited to, resistors, capacitors, etc. The target object is not limited to a substrate, and may be any workpiece that can be subjected to a predetermined processing.

[0053] During production, the mounting machine 200 periodically outputs equipment log data including multiple types of mounting logs to the production support device 100. Fig. 1 shows an example in which four types of mounting logs, a first mounting log L1, a second mounting log L2, a third mounting log L3, and a fourth mounting log L4, are output as equipment log data. Note that the number of types of mounting logs output by the mounting machine 200 is not limited to four, and may be two or more.

[0054] The mounting machine 200 may also have a sensor that measures the surrounding environment (for example, the temperature environment, the light environment, the air environment, etc.). The mounting machine 200 may have at least one of, for example, a temperature sensor that measures the air temperature or the temperature of a predetermined location on the mounting machine, a humidity sensor that measures humidity, an illuminance sensor that measures illuminance, and a dust sensor that measures airborne dust. The sensor that measures the surrounding environment may be located inside the mounting machine 200 or may be installed outside the mounting machine 200. The sensor data that measures the surrounding environment is environmental data that indicates the environment around the mounting machine 200, and is an example of a mounting log.

[0055] The mounting machine 200 may also be configured to be able to store, as an event log, a log of equipment operations performed by an operator on the mounting machine 200. The equipment operations may be, for example, operations performed by an operator via an operation panel attached to the mounting machine 200 or a PC (personal computer), and examples of such operations include operations to change the mounting conditions of the mounting machine 200 and operations to edit data. The equipment operations may also be, for example, tasks performed by an operator on the mounting machine 200, and examples of such tasks include model switching and material supply (e.g., tape splicing). The event log is an example of a mounting log.

[0056] The mounting machine 200 is also configured to be able to count the defect rate for each unit. Examples of defects include, but are not limited to, pickup errors when picking up a component, component recognition errors when recognizing a component, and component placement errors when placing a component. Taking pickup errors as an example, the pickup error rate can be calculated by dividing the number of pickup errors by the number of pickup attempts. The occurrence of a pickup error can be detected, for example, by changes in the flow rate of suction air passing through the nozzle hole of the nozzle, or by an image of a nozzle that has completed the component pickup operation (e.g., that should have picked up a component). The defect rates for pickup errors, component recognition errors, placement errors, etc. are examples of mounting logs.

[0057] The mounting machine 200 is also configured to be able to store data related to defects in mounting as quality data. Defects include, but are not limited to, mounting defects. Mounting defects include misalignment of mounting positions, incorrect mounting, and no mounting. The mounting machine 200 may store, for example, a mounting defect rate as quality data. The mounting defect rate can be calculated by dividing the number of mounting errors by the number of mounting attempts. The occurrence of mounting errors can also be obtained, for example, by capturing an image of the board. The quality data is an example of a mounting log.

[0058] The mounting machine 200 has multiple sensors that measure the status of each part of the unit. The multiple sensors measure the status of each part, for example, the drive control unit, the component mounting head unit (head), and the component supply unit (feeder). Examples of the multiple sensors include, but are not limited to, a flow sensor that measures the flow rate and is provided in a flow path that supplies air to the suction holes of a nozzle having suction holes for component suction, a current sensor that measures the current value of a motor that drives the feeder, a tension sensor that measures the tension applied to a cover tape attached to a base tape on which components are stored when the feeder is a tape feeder, and a camera that captures recognition images for component recognition. Note that the sensor data acquired by the sensors that measure the status of the mounting machine 200 is an example of a mounting log.

[0059] The multiple types of mounting logs may include at least two of the environmental data, event log, defect rate, quality data, and sensor data, and may also include types of mounting logs other than those listed above.

[0060] Note that the mounting log may also include, for example, information (e.g., pickup correction amount and recognition correction amount) regarding control amounts (correction amounts) for controlling the movement and rotation of a unit (e.g., a nozzle). The recognition correction amount indicates the amount of correction to the position of the nozzle in the X-axis and Y-axis directions after the nozzle picks up a component. The recognition correction amount indicates, for example, the amount of adjustment to the nozzle's position in the X-axis and Y-axis directions to mount the component on a mounting target position on the board. For example, the recognition correction amount is a value based on the difference between the position where the nozzle picks up the component and the pickup target position.

[0061] The pickup correction amount indicates the amount of correction made to the nozzle position in the X-axis and Y-axis directions when the nozzle picks up a component. The pickup correction amount indicates, for example, the amount of adjustment made when adjusting the nozzle position so that the nozzle can pick up the component's pickup target position (e.g., the center position on the top surface). For example, the pickup correction amount is a value based on the difference between the pickup target position of a component supplied by a feeder and the pickup center of the nozzle.

[0062] The display device 300 is, for example, a liquid crystal display device, and displays possible causes of malfunction in the mounting machine 200. The display device 300 may be a portable display device such as a smartphone, or may be a stationary display device. The display device 300 is an example of a display unit.

[0063] Here, the production support device 100 will be further described with reference to Figures 2 to 5. Figures 2 to 5 are diagrams showing the configuration of each estimation processing unit according to this embodiment. Note that, as shown in Figures 2 to 5, each estimation processing unit is shown to have a storage unit, but a common storage unit may also be used.

[0064] As shown in FIG. 1 , the production support device 100 has, as its functional configuration, an analysis unit 100a, a comprehensive judgment unit 50, an acquisition unit 60, and an output unit 70. The production support device 100 is realized, as its hardware configuration, by a non-volatile memory in which a program is stored, a volatile memory that is a temporary storage area for executing the program, an input / output port, a communication interface, a processor that executes the program, and the like. The analysis unit 100a, the comprehensive judgment unit 50, the acquisition unit 60, and the output unit 70 are realized by a processor that executes a program stored in memory, and the like. The production support device 100 may be realized by a stationary personal computer (PC), a mobile terminal such as a smartphone or tablet, a dedicated computer, or the like, or by a server (e.g., a cloud server), or by a combination thereof.

[0065] The analysis unit 100a is a processing unit that uses multiple analysis engines, each of which analyzes multiple types of implementation logs, to analyze multiple malfunction factors related to the malfunction of the mounting machine 200 in each of the multiple analysis engines. The analysis unit 100a includes a first estimation processing unit 10 to a fourth estimation processing unit 40 as multiple analysis engines. For example, different types of implementation logs are input to each of the first estimation processing unit 10 to the fourth estimation processing unit 40. That is, the implementation log input to the first estimation processing unit 10, the implementation log input to the second estimation processing unit 20, the implementation log input to the third estimation processing unit 30, and the implementation log input to the fourth estimation processing unit 40 are all different types. In the example of FIG. 1 , the first estimation processing unit 10 receives the first implementation log L1, the second estimation processing unit 20 receives the second implementation log L2, the third estimation processing unit 30 receives the third implementation log L3, and the fourth estimation processing unit 40 receives the fourth implementation log L4. In other words, the second mounting log L2 to the fourth mounting log L4 of the equipment log data are not input to the first estimation processing unit 10, the first mounting log L1, the third mounting log L3, and the fourth mounting log L4 of the equipment log data are not input to the second estimation processing unit 20, the first mounting log L1, the second mounting log L2, and the fourth mounting log L4 of the equipment log data are not input to the third estimation processing unit 30, and the first mounting log L1 to the third mounting log L3 of the equipment log data are not input to the fourth estimation processing unit 40. The first mounting log L1 to the fourth mounting log L4 are different types of mounting logs.

[0066] Furthermore, for example, mounting logs acquired during the same period may be input to each of the first estimation processing unit 10 to the fourth estimation processing unit 40, or mounting logs acquired at least partially during different periods may be input to each of the first estimation processing unit 10 to the fourth estimation processing unit 40. The number of estimation processing units (e.g., analysis engines) included in the production support device 100 is not particularly limited, and may be two or more.

[0067] The first estimation processing unit 10 receives an input of the first mounting log L1 and estimates the possibility of two or more malfunction factors of the mounting machine 200 corresponding to the first mounting log L1 from among multiple malfunction factors that cause the mounting machine 200 to malfunction. The first estimation processing unit 10 receives an input of the first mounting log L1 and estimates the possibility of two or more malfunction factors of the mounting machine 200 using a machine learning model that has been trained in advance by machine learning so as to output the possibility of two or more malfunction factors. Two or more malfunction factors of the mounting machine 200 corresponding to the first mounting log L1 mean two or more malfunction factors of the mounting machine 200 that can be identified from the first mounting log L1. The same applies to similar expressions hereinafter.

[0068] As shown in FIG. 2 , in this embodiment, the first estimation processing unit 10 is a processing unit that analyzes environmental factors and 5M (Machine / Material / Human / Measurement / Method) factors among multiple malfunction factors, and estimates where the cause is from an environmental / 5M perspective (whether it is an environmental factor or a 5M factor). The first estimation processing unit 10 includes a first environmental / 5M factor estimation unit 11, a second environmental / 5M factor estimation unit 12, and a first storage unit 13. The first mounting log L1 includes a mounting log that allows for analysis of the environmental factors and the 5M factors. In this embodiment, the first mounting log L1 includes a temperature L1a1 and a humidity L1b1.

[0069] The first environmental / 5M factor estimation unit 11 and the second environmental / 5M factor estimation unit 12 receive different mounting logs as input and use machine learning models trained to output the probabilities of the environmental factors and 5M factors to analyze the environmental factors and 5M factors related to a malfunction of the mounting machine 200. In this embodiment, the first environmental / 5M factor estimation unit 11 receives temperature L1a1 as input and uses a machine learning model trained to output the probabilities of the environmental factors and 5M factors to analyze the environmental factors and 5M factors related to a malfunction of the mounting machine 200. Furthermore, the second environmental / 5M factor estimation unit 12 receives humidity L1b1 as input and uses a machine learning model trained to output the probabilities of the environmental factors and 5M factors to analyze the environmental factors and 5M factors related to a malfunction of the mounting machine 200.

[0070] The first environmental / 5M factor estimation unit 11 may estimate the possibility of the environmental factors and the 5M factors by inputting the temperature L1a1 at a certain point in time into a machine learning model, or may estimate the possibility of the environmental factors and the 5M factors by inputting time series data of the temperature L1a1 into the machine learning model. The same applies to the second environmental / 5M factor estimation unit 12, and time series data of the humidity L1b1 may be input into the machine learning model.

[0071] An environment / 5M factor estimation unit is provided for each mounting log (here, temperature L1a1 and humidity L1b1) input to the first estimation processing unit 10. In this embodiment, one environment / 5M factor estimation unit is configured for temperature L1a1, and one environment / 5M factor estimation unit is configured for humidity L1b1.

[0072] Each of the first environment / 5M factor estimation unit 11 and the second environment / 5M factor estimation unit 12 is an analysis engine that is trained to specialize in determining whether the cause of a malfunction is the environment or the 5M. For example, the second implementation log L2 to the fourth implementation log L4 do not need to be used in the training of the first environment / 5M factor estimation unit 11 and the second environment / 5M factor estimation unit 12.

[0073] The first storage unit 13 is a storage device that stores the estimation results L1a2 and L1b2 of each factor estimation unit of the first estimation processing unit 10. The first storage unit 13 is realized, for example, by a semiconductor memory or the like, but is not limited to this. Each of the estimation results L1a2 and L1b2 includes a probability indicating the possibility that the environmental factor and the 5M factor are the cause of the malfunction.

[0074] 1 , the second estimation processing unit 20 receives the second mounting log L2 and estimates the possibility of two or more malfunction factors of the mounting machine 200 corresponding to the second mounting log L2 from among multiple malfunction factors that cause the malfunction of the mounting machine 200. The second estimation processing unit 20 receives the second mounting log L2 and estimates the possibility of two or more malfunction factors of the mounting machine 200 using a machine learning model that has been trained in advance by machine learning so as to output the possibility of two or more malfunction factors.

[0075] As shown in FIG. 3 , the second estimation processing unit 20 is a processing unit that analyzes factors of the multiple units that make up the mounting machine 200, among the multiple malfunction factors, and estimates the likelihood of each unit in the mounting machine 200 being the malfunction factor. The second estimation processing unit 20 includes a first unit factor estimation unit 21, a second unit factor estimation unit 22, and a second storage unit 23. The second mounting log L2 includes a mounting log that allows for analysis of unit factors. In this embodiment, the second mounting log L2 includes a defect rate for each unit, such as a pickup error rate L2a1 and a component recognition error rate L2b1. Examples of the multiple units include, but are not limited to, a head and a feeder.

[0076] The first unit factor estimation unit 21 and the second unit factor estimation unit 22 receive different mounting logs as input and use machine learning models trained to output the likelihood of each unit factor to analyze unit factors related to the malfunction of the mounting machine 200. In the present embodiment, the first unit factor estimation unit 21 receives the pickup error rate L2a1 as input and uses a machine learning model trained to output the likelihood of the unit factor to analyze unit factors related to the malfunction of the mounting machine 200. Furthermore, the second unit factor estimation unit 22 receives the component recognition error rate L2b1 as input and uses a machine learning model trained to output the likelihood of the unit factor to analyze unit factors related to the malfunction of the mounting machine 200.

[0077] The first unit factor estimation unit 21 may estimate the possibility of a unit factor by inputting the pickup error rate L2a1 at a certain point in time into a machine learning model, or may estimate the possibility of a unit factor by inputting time series data of the pickup error rate L2a1 into the machine learning model. The same applies to the second unit factor estimation unit 22, and time series data of the component recognition error rate L2b1 may be input into the machine learning model.

[0078] A unit factor estimation section is provided in one-to-one correspondence with each mounting log (here, pickup error rate L2a1 and component recognition error rate L2b1) input to the second estimation processing section 20.

[0079] Each of the first unit factor estimation unit 21 and the second unit factor estimation unit 22 is an analysis engine trained to specialize in determining which unit is the cause of a malfunction. In the case of a pickup error rate, factors related to pickup errors include the head, feeder, and component. The first unit factor estimation unit 21 may be, for example, an analysis engine trained to specialize in determining whether the cause of a malfunction is the head, feeder, or component. For example, the first unit factor estimation unit 21 and the second unit factor estimation unit 22 may not use the first mounting log L1, the third mounting log L3, and the fourth mounting log L4 for training.

[0080] The second storage unit 23 is a storage device that stores the estimation results L2a2 and L2b2 of the factor estimation units of the second estimation processing unit 20. The second storage unit 23 is realized by, for example, a semiconductor memory or the like, but is not limited to this.

[0081] 1 , the third estimation processing unit 30 receives the third mounting log L3 and estimates the possibility of two or more malfunction factors of the mounting machine 200 corresponding to the third mounting log L3 from among multiple malfunction factors that cause the malfunction of the mounting machine 200. The third estimation processing unit 30 receives the third mounting log L3 and estimates the possibility of two or more malfunction factors of the mounting machine 200 using a machine learning model that has been trained in advance by machine learning so as to output the possibility of two or more malfunction factors.

[0082] As shown in FIG. 4 , the third estimation processing unit 30 is a processing unit that analyzes factors related to the worker performing the work on the mounting machine 200 among multiple malfunction factors and estimates the factor from among the possible human work factors. The third estimation processing unit 30 includes a first human factor estimation unit 31, a second human factor estimation unit 32, and a third storage unit 33. The third mounting log L3 includes a mounting log that allows analysis of human work factors. In this embodiment, the third mounting log L3 includes an event log L3a1 for each manual work and a defect rate L3b1 before and after the event. For example, the event log L3a1 is an event log related to manual tape splicing work, and the defect rate L3b1 before and after the event is the defect rate (e.g., pickup error rate) before and after tape splicing.

[0083] The first human factors estimation unit 31 and the second human factors estimation unit 32 receive input of different mounting logs and use machine learning models trained to output the likelihood of each human work factor to analyze human work factors related to the malfunction of the mounting machine 200. In the present embodiment, the first human factors estimation unit 31 receives input of the event log L3a1 and uses a machine learning model trained to output the likelihood of the human work factor to analyze human work factors related to the malfunction of the mounting machine 200. Furthermore, the second human factors estimation unit 32 receives input of the defect rate L3b1 before and after the event and uses a machine learning model trained to output the likelihood of the human work factor to analyze human work factors related to the malfunction of the mounting machine 200.

[0084] The human factors estimation unit is provided in one-to-one correspondence with each mounting log (here, the event log L3a1 and the defect rate L3b1 before and after the event) input to the third estimation processing unit 30.

[0085] Each of the first human factors estimation unit 31 and the second human factors estimation unit 32 is an analysis engine that has been trained to specialize in distinguishing between human work and other factors (e.g., equipment factors, component factors, environmental factors). For example, the first mounting log L1, the second mounting log L2, and the fourth mounting log L4 do not need to be used in the training of the first human factors estimation unit 31 and the second human factors estimation unit 32.

[0086] The third storage unit 33 is a storage device that stores the estimation results L3a2 and L3b2 of the factor estimation units of the third estimation processing unit 30. The third storage unit 33 is realized by, for example, a semiconductor memory or the like, but is not limited to this.

[0087] 1 , the fourth estimation processing unit 40 receives the fourth mounting log L4 and estimates the possibility of two or more malfunction factors of the mounting machine 200 corresponding to the fourth mounting log L4 from among multiple malfunction factors that cause the malfunction of the mounting machine 200. The fourth estimation processing unit 40 receives the fourth mounting log L4 and estimates the possibility of two or more malfunction factors of the mounting machine 200 using a machine learning model that has been trained in advance by machine learning so as to output the possibility of two or more malfunction factors.

[0088] 5 , the fourth estimation processing unit 40 is a processing unit that analyzes, among multiple malfunction factors, factors of each part that constitutes a unit of the mounting machine 200, and estimates the factors of each part in the unit. The fourth estimation processing unit 40 includes a first unit / part factor estimation unit 41, a second unit / part factor estimation unit 42, and a fourth storage unit 43. The fourth mounting log L4 includes a mounting log that allows analysis of the factors of each part. In this embodiment, the fourth mounting log L4 includes a flow rate L4a1 measured by a flow sensor provided in the nozzle flow path and a current value L4b1 of the motor of the feeder.

[0089] The first unit / parts factor estimation unit 41 and the second unit / parts factor estimation unit 42 receive different types of mounting logs as input, and use machine learning models trained to output the likelihood that a part of a unit is the cause of the malfunction if that unit is the cause of the malfunction, to analyze part factors related to the malfunction of the mounting machine 200. In this embodiment, the first unit / parts factor estimation unit 41 receives input of the flow rate L4a1, and uses a machine learning model trained to output the likelihood of a part factor of a certain unit, to analyze part factors related to the malfunction of the mounting machine 200. Furthermore, the second unit / parts factor estimation unit 42 receives input of the motor current value L4b1, and uses a machine learning model trained to output the likelihood of a part factor of another unit, to analyze part factors related to the malfunction of the mounting machine 200.

[0090] The first unit / parts factor estimation unit 41 and the second unit / parts factor estimation unit 42 each estimate factors using a machine learning model trained specifically for a specific unit. For example, if the units are a head and a feeder, the first unit / parts factor estimation unit 41 can use a machine learning model trained specifically for the head, and the second unit / parts factor estimation unit 42 can use a machine learning model trained specifically for the feeder. For example, the first unit / parts factor estimation unit 41 estimates the probability of factors for multiple parts constituting the head, and the second unit / parts factor estimation unit 42 estimates the probability of factors for multiple parts constituting the feeder. Examples of parts constituting the head include a primary filter and a secondary filter arranged in a nozzle flow path. Note that the first unit / parts factor estimation unit 41 and the second unit / parts factor estimation unit 42 may each be configured to estimate the probability of factors for each part for one unit, for different units, or for each part for multiple units.

[0091] In addition to the flow rate L4a1 and the motor current value L4b1, the fourth estimation processing unit 40 may use, for example, air temperature as input data, as in the first estimation processing unit 10. Part of the implementation log input to the estimation processing unit may be common to the implementation log input to the other estimation processing unit. Note that the two estimation processing units share common input data during learning but operate using different machine learning models because they use different ground truth data.

[0092] The unit / part factor estimation section is provided for each mounting log (here, flow rate L4a1 and motor current value L4b1) input to the fourth estimation processing section 40, on a one-to-one basis.

[0093] The first unit / parts factor estimation unit 41 and the second unit / parts factor estimation unit 42 are each an analysis engine that has been trained to specifically identify which part of the unit is the cause of a malfunction. For example, the second mounting log L2 and the third mounting log L3 do not need to be used in the training of the first unit / parts factor estimation unit 41 and the second unit / parts factor estimation unit 42.

[0094] The fourth storage unit 43 is a storage device that stores the estimation results L4a2 and L4b2 of the factor estimation units of the fourth estimation processing unit 40. The fourth storage unit 43 is realized by, for example, a semiconductor memory or the like, but is not limited to this.

[0095] In the above-described production support device 100, each estimation processing unit (i.e., each analysis engine) may be configured to be able to analyze multiple types of implementation logs in parallel. Being able to analyze multiple types of implementation logs in parallel may mean that different types of implementation logs are input to each estimation processing unit and analyzed, or that multiple types of implementation logs are analyzed in parallel over time.

[0096] Although the example has been described in which each of the first estimation processing unit 10 to the fourth estimation processing unit 40 has two factor estimation units, the number of factor estimation units may be one or more. Furthermore, the first estimation processing unit 10 to the fourth estimation processing unit 40 may have the same number of estimation processing units or different numbers of estimation processing units.

[0097] The production support device 100 may also include a pre-processing unit that sorts the equipment log data acquired from the mounting machine 200 into input data for each estimation processing unit and outputs the sorted mounting log to each estimation processing unit.

[0098] 1 , the comprehensive judgment unit 50 estimates the cause of the malfunction of the mounting machine 200 based on the estimation results of each estimation processing unit. The comprehensive judgment unit 50 estimates the possibility of each factor causing the malfunction of the mounting machine 200, for example, based on the estimation results of each estimation processing unit. The factors causing the malfunction include at least two of equipment factors indicating factors related to the equipment, work factors indicating factors related to the work performed by the worker, part factors indicating factors related to the parts used, and environmental factors indicating factors related to the surrounding environment. The processing of the comprehensive judgment unit 50 will be described later with reference to FIG. 7 .

[0099] The acquisition unit 60 is a communication interface that allows the production support device 100 to communicate with the mounting machine 200, and acquires equipment log data from the mounting machine 200. The acquisition unit 60 may acquire the first mounting log L1 to the fourth mounting log L4 as equipment log data from the mounting machine 200, or may acquire the data via another device. The acquisition unit 60 is configured to include, for example, a communication circuit (or a communication module), but is not limited to this.

[0100] In addition, the acquisition unit 60 outputs the acquired first implementation log L1 to the first estimation processing unit 10, outputs the acquired second implementation log L2 to the second estimation processing unit 20, outputs the acquired third implementation log L3 to the third estimation processing unit 30, and outputs the acquired fourth implementation log L4 to the fourth estimation processing unit 40.

[0101] The output unit 70 is a communication interface for the production support device 100 to communicate with the display device 300, and outputs the estimation results of the likelihood of each cause of the malfunction of the mounting machine 200 to the display device 300. The output unit 70 is configured to include, for example, a communication circuit (or a communication module), but is not limited to this.

[0102] 2. Operation of the Production Support Device Next, the operation of the production support device configured as described above will be described with reference to Fig. 6 and Fig. 7. Fig. 6 is a flowchart showing the operation (production support method) of the production support device 100 according to this embodiment.

[0103] 6, first, the acquisition unit 60 of the production support device 100 acquires various mounting logs from the mounting machine 200 (S11). The acquisition timing of each of the first mounting log L1 to fourth mounting log L4 may be the same or different.

[0104] Next, the first estimation processing unit 10 estimates environmental and 5M factors from among the multiple malfunction factors based on the first implementation log L1 (S12). For example, the first estimation processing unit 10 obtains the factor probabilities of environmental factors and other factors, which are the output of a machine learning model obtained by inputting environmental data (e.g., temperature L1a1) into the machine learning model. For example, the sum of the factor probabilities of environmental factors and other factors is 100%.

[0105] Next, the second estimation processor 20 estimates a unit factor among the multiple malfunction factors based on the second mounting log L2 (S13). For example, the second estimation processor 20 obtains the factor probabilities of equipment factors (heads and feeders) and part factors, which are the output of a machine learning model obtained by inputting the defect rate (e.g., pickup error rate L2a1) into the machine learning model. For example, the sum of the factor probabilities of equipment factors (heads and feeders) and part factors is 100%.

[0106] Next, the third estimation processor 30 estimates a human factor among the multiple malfunction factors based on the third mounting log L3 (S14). For example, the third estimation processor 30 obtains the factor probabilities of equipment factors, work factors, and part factors, which are the output of a machine learning model obtained by inputting the event log L3a1 into the machine learning model. For example, the sum of the factor probabilities of equipment factors, work factors, and part factors may be 100%, for example.

[0107] Next, the fourth estimation processing unit 40 estimates a unit / part factor from among the multiple malfunction factors based on the fourth mounting log L4 (S15). For example, the fourth estimation processing unit 40 obtains the factor probability of each part (or the part camera in the case of a part camera) among the equipment factors, which is the output of the machine learning model obtained by inputting the flow rate L4a1 into the machine learning model. For example, the sum of the factor probability of each part and the factor probability of the part camera is 100%.

[0108] Furthermore, a fifth estimation processing unit (not shown) of the production support device 100 may estimate a quality factor from among the multiple malfunction factors based on the fifth mounting log. For example, the fifth estimation processing unit obtains the factor probabilities of equipment factors (heads and feeders), work factors, and component factors, which are the output of a machine learning model obtained by inputting the mounting defect rate into the machine learning model. For example, the sum of the factor probability of equipment factors (heads and feeders), the factor probability of work factors, and the factor probability of component factors is 100%.

[0109] Next, the comprehensive judgment unit 50 makes a comprehensive judgment of the cause of the malfunction of the mounting machine 200 based on each of the factors estimated in steps S12 to S15 (S16). It can also be said that the comprehensive judgment unit 50 judges the cause of the malfunction of the mounting machine 200 using the estimated result of the malfunction as an explanatory variable and the possibility of the final cause as a target variable.

[0110] Here, the overall judgment process by the overall judgment unit 50 will be described with reference to Fig. 7. Fig. 7 is a diagram showing an example of the display of the overall judgment result of the production support device 100 according to this embodiment. Note that Fig. 7 shows the overall judgment result displayed on the display device 300.

[0111] FIG. 7 shows the overall judgment results when the production support device 100 is equipped with five estimation processing units, i.e., five analysis engines. Estimate I shows the output result of the first estimation processing unit 10 when environmental data is input to the first estimation processing unit 10 as a mounting log. Estimate II shows the output result of the third estimation processing unit 30 when the event log L3a1 is input to the third estimation processing unit 30 as a mounting log. Estimate III shows the output result of the second estimation processing unit 20 when the defect rate is input to the second estimation processing unit 20 as a mounting log. Estimate IV shows the output result of the fifth estimation processing unit when quality data is input to the fifth estimation processing unit as a mounting log. Estimate V shows the output result of the fourth estimation processing unit 40 when sensor data is input to the fourth estimation processing unit 40 as a mounting log.

[0112] The multiple malfunction factors (candidate factors) include equipment factors, work factors, part factors, and the environment (environmental factors). The equipment factors include, as units, a head, a feeder, and a part camera. The head includes, as parts, a primary filter and a secondary filter provided in the flow path, and the feeder includes, as parts, conveying surface contamination and motor deterioration. Conveying surface contamination refers to contamination on the surface of the cover tape used to pack parts into the pockets of the carrier tape in the tape feeder, and motor deterioration refers to deterioration of the motor that rotates the feeder. Motor deterioration can be estimated by measuring the motor current value L4b1. If the event log L3a1 includes information that a worker is performing tape splicing work, the work factor is splicing. The part factor is the part, and the environment is temperature.

[0113] Note that "-" in FIG. 7 means that the analysis engine does not output the possibility for that cause candidate.

[0114] As shown in Figure 7, it is difficult to estimate detailed factors using environmental data other than the environment. Furthermore, other implementation logs output only a portion of the possible factors. Thus, when using only a single analysis engine, it may not be possible to accurately estimate factors.

[0115] On the other hand, in this embodiment, multiple estimation processing units (multiple analysis engines) are used, so even if there is a candidate factor for which the estimation processing unit cannot output the possibility of the factor, this can be compensated for by the output of another estimation processing unit.

[0116] For example, one of the multiple estimation processing units receives a log of a corresponding type as input and calculates the factor probability of each of two or more first malfunction factors corresponding to that type of log among the multiple malfunction factors. Also, another of the multiple estimation processing units receives a log of a corresponding type as input and calculates the factor probability of each of two or more second malfunction factors corresponding to the second type of log among the multiple malfunction factors, the second malfunction factors being at least partially different from the two or more first malfunction factors. This makes it possible to compensate for the factor probability of a factor not output in the estimation result of one estimation processing unit with the estimation result of the other estimation processing unit.

[0117] Furthermore, by having multiple analysis engines, the amount of processing is expected to be reduced compared to when a single large-scale analysis engine is used. In other words, the production support device 100 according to this embodiment can improve the accuracy of the causes of malfunctions in production equipment while reducing the processing load.

[0118] The comprehensive evaluation unit 50 calculates the probability of each candidate factor based on the estimation results I to V shown in FIG. 7 . The comprehensive evaluation unit 50 calculates the probability that the primary filter of the equipment head is the cause of the malfunction using the outputs of Estimates I to V, which are 90%, 20%, 70%, 50%, and 20%, respectively. The comprehensive evaluation unit 50 may, for example, input the type of mounting data and the probability (90%, 20%, 70%, 50%, and 20%) into a machine learning model to obtain the output of the comprehensively determined probability of the malfunction factor as a value indicating the probability of the malfunction factor. Alternatively, the comprehensive evaluation unit 50 may obtain the value obtained by multiplying the five probabilities (e.g., multiplying using weights) as a value indicating the probability of the malfunction factor. The calculated probability of the malfunction is information indicating the priority among multiple candidate malfunction locations. The probability of the malfunction is an example of the possibility of the cause of the malfunction of the mounting machine 200.

[0119] Furthermore, when the candidate factor is a component camera, the analysis engine output for the defect rate and quality data is not obtained ("-" in FIG. 7 ). Therefore, the probability of the cause may be calculated using the other three probabilities, or a numerical value may be assigned to the "-" to calculate the probability of the cause. Taking the example of a case where the candidate factor is a component camera, if there is a difference between the probability of the cause estimated using 90%, 20%, and 60% during machine learning model training and the correct data, a "-" numerical value may be preset to narrow the difference. Furthermore, the "-" numerical value may be, for example, a fixed value. In this way, the comprehensive judgment unit 50 may calculate the probability of the cause based on the probability of the first factor calculated by the analysis engine and the preset probability corresponding to the analysis engine.

[0120] 6 again, next, the output unit 70 outputs the overall judgment result to the display device 300 (S17). The output unit 70 outputs the overall judgment result including the calculated possibility of the cause of the malfunction of the mounting machine 200 for each of the multiple malfunction factors to the display device 300, thereby causing the display device 300 to display the overall judgment result. The output unit 70 may, for example, output the overall judgment result including the table showing the relationship between the factors and the logs shown in FIG. 7 to the display device 300, causing the table showing the relationship between the factors and the logs shown in FIG. 7 to be displayed on the display device 300.

[0121] The comprehensive evaluation unit 50 may estimate the probability of each candidate cause and display an illustration (sun, cloud, umbrella in the example of FIG. 7 ) corresponding to the numerical value of the estimated probability of the cause next to the numerical value via the output unit 70. This allows the mounting line manager or the like to intuitively know the degree of probability of the malfunction by looking at the illustration, and to grasp the priority of the response for each candidate cause. In other words, the production support device 100 can support the manager or the like in dealing with malfunctions in the mounting machine.

[0122] The comprehensive judgment unit 50 may also display the results of the malfunction timing prediction. The comprehensive judgment unit 50 may predict the timing of a malfunction based on time-series data from one sensor. For example, the comprehensive judgment unit 50 may predict the timing of a malfunction based on the timing at which the value of the time-series data exceeds a predetermined value. The time-series data may include, for example, data measured by a sensor before and after a person performs work (e.g., tape splicing).

[0123] 7, by displaying the results of each analysis engine and the results of the overall judgment unit 50 using those results in a single table, it is possible to visualize a plurality of malfunction factor estimation results at a bird's-eye view when making a comprehensive judgment on the causes of malfunction. By displaying such a table on the display device 300 and having a manager or the like check it, it is possible to effectively support making a comprehensive judgment using the results of each analysis engine that estimates factors.

[0124] The comprehensive judgment unit 50 may display at least the candidate factors and the possible causes shown in Fig. 7. The comprehensive judgment unit 50 may also extract the candidate factor with the highest possible cause or a probability equal to or greater than a predetermined value, and display it as the cause of the malfunction.

[0125] In the example display shown in FIG. 7, the secondary filter has a high probability of being the cause because each estimation result has a high probability. It can be seen from the sensor data that the secondary filter is more likely to be the cause of the malfunction than the primary filter. By viewing the estimation results from a bird's-eye view and comparing them with the results of other engines, the validity of the results can be easily understood. The table shown in FIG. 7 may also display the reason for the determination.

[0126] While the production support device according to one or more aspects has been described above based on the embodiments, the present disclosure is not limited to these embodiments. As long as it does not deviate from the spirit of the present disclosure, various modifications conceivable by a person skilled in the art to the present embodiments and embodiments constructed by combining components of different embodiments may also be included in the present disclosure.

[0127] For example, in the above embodiment, an example in which the production support device is used in a mounting machine (component mounting machine) has been described, but the production support device may also be used in production devices other than mounting machines. For example, the present disclosure may be realized as a production support device including: an acquisition unit that acquires multiple types of production logs related to the production device; multiple analysis engines that analyze the multiple types of production logs, each of which receives at least some different types of production logs; an analysis unit that analyzes multiple malfunction factors related to the production device using each of the multiple analysis engines; and a determination unit that calculates the likelihood of each of the multiple malfunction factors being the cause of the malfunction of the production device based on the analysis results of each of the multiple analysis engines. The production device may be a semiconductor manufacturing device that manufactures semiconductor components, a processing device that processes the shape of an object, or other device. Furthermore, the production log may be a type of log corresponding to the production device.

[0128] Furthermore, although the numerical values ​​described in the above embodiments are average values, they may be other statistical values, such as medians, modes, maximum values, and minimum values.

[0129] In the above embodiment, an example has been described in which each factor estimation unit estimates a malfunction factor using a machine learning model, but the use of a machine learning model is not limited thereto. Each factor estimation unit may estimate a malfunction factor based on, for example, a table in which implementation logs are associated with malfunction factors.

[0130] Furthermore, each machine learning model described in the above embodiments may use a neural network (including deep learning using a multi-layer neural network), genetic programming, a decision tree, a Bayesian network, a support vector machine (SVM), a Gaussian process, or the like.

[0131] Furthermore, in the above embodiment, an example has been described in which the production support device and the display device are separate entities, but this is not limiting, and for example, the production support device and the display device may be an integrated device.

[0132] In the above embodiments, each component may be configured with dedicated hardware, or may be realized by executing a software program suitable for each component. Each component may be realized by a program execution unit such as a CPU or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory.

[0133] The order in which the steps in the flowchart are executed is merely an example for specifically explaining the present disclosure, and other orders may be used. Some of the steps may be executed simultaneously (in parallel) with other steps, or some of the steps may not be executed.

[0134] The division of functional blocks in the block diagram is an example, and multiple functional blocks may be realized as a single functional block, one functional block may be divided into multiple blocks, or some functions may be moved to another functional block.Furthermore, the functions of multiple functional blocks having similar functions may be processed in parallel or in time-sharing by a single piece of hardware or software.

[0135] Furthermore, the production support device according to the above-described embodiment may be realized as a single device or may be realized by multiple devices. When the production support device is realized by multiple devices, the components of the production support device may be distributed in any manner among the multiple devices. When the production support device is realized by multiple devices, the communication method between the multiple devices is not particularly limited, and may be wireless communication or wired communication. Furthermore, wireless communication and wired communication may be combined between the devices.

[0136] Furthermore, each component described in the above embodiments may be implemented as software or, typically, as an LSI, which is an integrated circuit. These components may be individually integrated into a single chip, or some or all of them may be integrated into a single chip. Here, the term "LSI" is used, but depending on the level of integration, it may also be referred to as an IC, system LSI, super LSI, or ultra LSI. Furthermore, the integrated circuit implementation method is not limited to LSI, and may be implemented using a dedicated circuit (a general-purpose circuit that executes a dedicated program) or a general-purpose processor. After LSI fabrication, a field programmable gate array (FPGA) that can be programmed or a reconfigurable processor that can reconfigure the connections or settings of circuit cells within the LSI may also be used. Furthermore, if an integrated circuit technology that replaces LSI emerges due to advances in semiconductor technology or a derivative technology, that technology may naturally be used to integrate the components.

[0137] A system LSI is an ultra-multifunctional LSI manufactured by integrating multiple processing units on a single chip. Specifically, it is a computer system that includes a microprocessor, ROM (Read Only Memory), RAM (Random Access Memory), etc. Computer programs are stored in the ROM. The system LSI achieves its functions when the microprocessor operates in accordance with the computer program.

[0138] Furthermore, one aspect of the present disclosure may be a computer program that causes a computer to execute each of the characteristic steps included in the production support method shown in FIG. 6 .

[0139] Furthermore, for example, the program may be a program to be executed by a computer. Another aspect of the present disclosure may be a computer-readable non-transitory recording medium on which such a program is recorded. For example, such a program may be recorded on a recording medium and distributed or circulated. For example, the distributed program may be installed in a device having another processor, and the program may be executed by the processor, thereby causing the device to perform each of the above processes.

[0140] The present disclosure is useful for a monitoring device that monitors a mounting line, etc.

[0141] REFERENCE SIGNS LIST 1 Production support system 10 First estimation processing unit (analysis engine) 11 First environment / 5M factor estimation unit 12 Second environment / 5M factor estimation unit 13 First memory unit 20 Second estimation processing unit (analysis engine) 21 First unit factor estimation unit 22 Second unit factor estimation unit 23 Second memory unit 30 Third estimation processing unit (analysis engine) 31 First human factor estimation unit 32 Second human factor estimation unit 33 Third memory unit 40 Fourth estimation processing unit (analysis engine) 41 First unit / part factor estimation unit 42 Second unit / part factor estimation unit 43 Fourth memory unit 50 Overall judgment unit (judgment unit) 60 Acquisition unit 70 Output unit 100 Production support device 100a Analysis unit 200 Mounting machine 300 Display device (display unit) L1 First mounting log L1a1, L4a1 Flow rate (environmental data) L1a2, L1b2, L2a2, L2b2, L3a2, L3b2, L4a2, L4b2 Estimation results L1b1 Humidity (environmental data) L2 Second mounting log L2a1 Pickup error rate (defect rate) L2b1 Component recognition error rate (defect rate) L3 Third mounting log L3a1 Event log L3b1 Defect rate before and after the event L4 Fourth mounting log L4b1 Motor current value (sensor data)

Claims

1. A production support device comprising: an acquisition unit that acquires multiple types of mounting logs related to a mounting machine; an analysis unit having multiple analysis engines that analyze each of the multiple types of mounting logs, each of the multiple analysis engines analyzing multiple malfunction factors related to the mounting machine; and a determination unit that calculates the possibility that each of the multiple malfunction factors is the cause of the malfunction of the mounting machine based on the analysis results of each of the multiple analysis engines.

2. The production support device according to claim 1, wherein the plurality of types of mounting logs include at least two of environmental data, event logs, defect rates, quality data, and sensor data.

3. The production support device according to claim 1 or 2, wherein the plurality of malfunction factors include at least two of equipment factors, work factors, part factors, and environmental factors.

4. The production support device according to claim 1 or 2, wherein the determination unit causes a display unit to display the calculated probability of the cause of the malfunction of the mounting machine for each of the plurality of malfunction factors.

5. The production support device according to claim 4, wherein the determination unit causes the display unit to display the analysis results from each of the plurality of analysis engines as a table.

6. The production support device according to claim 1 or 2, wherein the plurality of analysis engines are configured to analyze the plurality of types of implementation logs in parallel.

7. The production support device according to claim 1 or 2, wherein each of the plurality of analysis engines is configured to analyze a different type of implementation log.

8. The production support device according to claim 1 or 2, wherein the plurality of analysis engines include a first analysis engine and a second analysis engine; the plurality of types of implementation logs include a first type log and a second type log that are different from each other; the first analysis engine receives the first type log as an input and analyzes each of two or more first malfunction factors that correspond to the first type log among the plurality of malfunction factors; and the second analysis engine receives the second type log as an input and analyzes each of two or more second malfunction factors that correspond to the second type log among the plurality of malfunction factors and that differ at least in part from the two or more first malfunction factors.

9. The production support device of claim 8, wherein the plurality of malfunction factors include a first malfunction factor and a second malfunction factor, the first malfunction factor is a factor corresponding to each of the first type of log and the second type of log, and the second malfunction factor is a factor corresponding to the first type of log, and the determination unit calculates the possibility of the first malfunction factor based on the possibility of the first malfunction factor calculated by the first analysis engine and the possibility of the first malfunction factor calculated by the second analysis engine, and calculates the possibility of the second malfunction factor based on the possibility of the first malfunction factor calculated by the first analysis engine.

10. The production support device described in claim 9, wherein the judgment unit calculates the possibility of the second malfunction factor based on the possibility of the first malfunction factor calculated by the first analysis engine and a possibility corresponding to the second analysis engine that is preset.

11. A production support method comprising: acquiring multiple types of mounting logs related to a mounting machine; having multiple analysis engines that analyze each of the multiple types of mounting logs; each of the multiple analysis engines analyzing multiple malfunction factors related to the mounting machine; and calculating the likelihood that each of the multiple malfunction factors is the cause of the malfunction of the mounting machine based on the results of the analysis by the multiple analysis engines.

12. A program for causing a computer to execute the production support method according to claim 11.

Images