Diagnostic system and diagnostic method

Abstract

A diagnostic system (100) is provided with: an acquisition unit (101) that acquires (i) production plan information (d2) indicating a production plan for producing, using production equipment, a mounted substrate which is a substrate (3) on which a component (D) is mounted, and (ii) production history information (d3) indicating the history of using the production equipment to produce mounted substrates; a prediction unit (103) that predicts an arrival time when the future state of degradation of a constituent member included in the production equipment reaches a prescribed degradation state, which is a predetermined degradation state, such prediction being on the basis of degradation information (db) indicating the past or current degradation state of the constituent member, and the production plan information (d2) and the production history information (d3) acquired by the acquisition unit (101); and an output unit (104) that outputs the arrival time information indicating the predicted arrival time.

Description

Technical Field

[0001] The present disclosure relates to a system for diagnosing production equipment such as a component mounting device.

Background Art

[0002] Conventionally, a component mounting device for determining whether a switching valve is normal has been proposed (see, for example, Patent Document 1). The component mounting device is production equipment for producing a mounting board on which components are mounted or attached to a substrate, and is also called a component mounting device. The switching valve is a component provided in the component mounting device, and selectively connects a vacuum pump and an air supply source to a suction nozzle. The suction nozzle is a nozzle that sucks and holds a component in order to attach the component to a substrate. When the vacuum pump is connected to the suction nozzle by the switching valve and the vacuum pump is driven, the suction nozzle sucks the surrounding air. On the other hand, when the air supply source is connected to the suction nozzle by the switching valve and the air supply source is driven, the suction nozzle blows out air to the surroundings. Such a component mounting device can be said to be equipped with a diagnostic system for determining whether a component, which is the switching valve, is normal.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the diagnostic system of Patent Document 1 has a problem that it is difficult to sufficiently support maintenance of production equipment by an operator.

[0005] Therefore, the present disclosure provides a diagnostic system and the like that can more appropriately support maintenance of production equipment by an operator. <头

Means for Solving the Problems

[0006] A diagnostic system according to one aspect of the present disclosure includes: an acquisition unit that acquires (i) production plan information indicating a production plan for producing a mounting board, which is a substrate on which components are mounted, using production equipment; and (ii) production performance information indicating the actual production of the mounting board using the production equipment; deterioration information indicating the past or present deterioration state of components included in the production equipment; a prediction unit that predicts the arrival time when the future deterioration state of the components will reach a predetermined deterioration state, which is a predetermined deterioration state, based on the production plan information and production performance information acquired by the acquisition unit; and an output unit that outputs arrival time information indicating the predicted arrival time.

[0007] These comprehensive or specific embodiments may be implemented as a system, method, integrated circuit, computer program, or recording medium such as a computer-readable CD-ROM (Compact Disc Read-Only Memory), or as any combination of a system, method, integrated circuit, computer program, and recording medium. Furthermore, the recording medium may be a non-temporary recording medium. [Effects of the Invention]

[0008] The diagnostic system described herein can better support workers in maintaining production equipment.

[0009] Further advantages and effects of one aspect of this disclosure will be made apparent from the specification and drawings. Such advantages and / or effects are provided by several embodiments and features described in the specification and drawings, but not all of them are necessarily provided in order to obtain one or more identical features. [Brief explanation of the drawing]

[0010] [Figure 1] Figure 1 is a plan view of a component mounting device in an embodiment. [Figure 2]Figure 2 is a perspective view of the transfer head used in the component mounting device according to the embodiment. [Figure 3] Figure 3 shows an example of the configuration of the air control mechanism in the embodiment. [Figure 4] Figure 4 is a block diagram showing an example of the functional configuration of a component mounting device in an embodiment. [Figure 5] Figure 5 is a block diagram showing an example of the functional configuration of the diagnostic system in the embodiment. [Figure 6] Figure 6 is a diagram illustrating the learning unit and diagnostic model in the embodiment. [Figure 7] Figure 7 is a diagram illustrating the deterioration identification part and diagnostic model in the embodiment. [Figure 8] Figure 8 shows a specific example of degradation information in the embodiment. [Figure 9] Figure 9 is a diagram illustrating the input and output of the abnormality processing unit, classification processing unit, and degradation degree processing unit in the embodiment. [Figure 10] Figure 10 is a diagram illustrating the processing of the prediction unit in the embodiment. [Figure 11] Figure 11 is a flowchart showing an example of the processing operation related to the identification of the deterioration state by the diagnostic system in the embodiment. [Figure 12] Figure 12 is a flowchart showing an example of the processing operation related to maintenance instructions by the diagnostic system in the embodiment. [Figure 13] Figure 13 is a flowchart showing an example of the processing operation related to maintenance timing prediction by the diagnostic system in the embodiment. [Modes for carrying out the invention]

[0011] Furthermore, the diagnostic system according to the first aspect of this disclosure includes: an acquisition unit that acquires (i) production plan information indicating a production plan for producing a mounting board, which is a substrate on which components are mounted, using production equipment; and (ii) production performance information indicating the actual production of the mounting board using the production equipment; deterioration information indicating the past or present deterioration state of components included in the production equipment; a prediction unit that predicts the arrival time when the future deterioration state of the components will reach a predetermined deterioration state, which is a specified deterioration state, based on the production plan information and production performance information acquired by the acquisition unit; and an output unit that outputs arrival time information indicating the predicted arrival time. For example, in the diagnostic system according to the seventh aspect which is dependent on the first aspect and any one of the second to sixth aspects described later, the specified deterioration state may be a first specified deterioration state in which maintenance of the components is required, or a second specified deterioration state in which preparation for the maintenance is required. Furthermore, in a diagnostic system according to an eighth embodiment dependent on the seventh embodiment, the prediction unit may estimate a future degree of deterioration, which is a value that increases as the degree of deterioration of the component increases, as a future deterioration state of the component, and predict, based on the production plan information, the time when the future degree of deterioration reaches a first threshold corresponding to the first defined deterioration state, or the time when it reaches a second threshold corresponding to the second defined deterioration state.

[0012] This allows the system to predict when maintenance of deteriorated components will be necessary, or when preparations for such maintenance will be required, and outputs information indicating the arrival date. Therefore, workers using the production equipment can understand the arrival date, i.e., the maintenance implementation date or maintenance preparation date. As a result, workers can perform their work in anticipation of these dates, improving work efficiency. It also allows for more efficient maintenance. In this way, it is possible to more appropriately support workers in performing maintenance on production equipment.

[0013] In addition, in the diagnostic system according to the second aspect dependent on the first aspect, the deterioration information may indicate the deterioration state of the component at each of a plurality of past time points.

[0014] Thereby, based on the change over time of the deterioration state, the future deterioration state can be appropriately estimated, and as a result, the prediction accuracy of the arrival time can be improved.

[0015] In addition, in the diagnostic system according to the third aspect dependent on the first aspect or the second aspect, the production facility is a component mounting device that adsorbs components by a transfer head and mounts them on a substrate, and the diagnostic system further includes a deterioration specifying unit that specifies the deterioration state of the component included in the transfer head based on flow rate information regarding the flow rate of air flowing through the transfer head, and the prediction unit may acquire information indicating the deterioration state specified by the deterioration specifying unit as the deterioration information.

[0016] Thereby, the deterioration state of the component included in the transfer head is specified based on the flow rate information, and the specified deterioration state is used for predicting the arrival time. Therefore, the deterioration state can be appropriately specified, and the prediction accuracy of the arrival time can be improved.

[0017] In addition, in the diagnostic system according to the fourth aspect dependent on the third aspect, the deterioration specifying unit may specify the deterioration state of the component by estimating a degree of deterioration indicating the degree of deterioration of the component.

[0018] Thereby, since the deterioration state is estimated in more detail as the degree of deterioration, the prediction accuracy of the arrival time can be further improved. [[ID=...]]

[0019] [[ID=...]] In addition, in the diagnostic system according to the fifth aspect dependent on the fourth aspect, the deterioration specifying unit may determine whether there is an abnormality in the component, and when it is determined that there is an abnormality, estimate the degree of deterioration of the component. <><end>

[0020] This allows for the estimation of the degree of deterioration for components determined to be abnormal. As a result, the time to reach deterioration can be predicted for components that are expected to show signs of deterioration, while the prediction of the time to reach deterioration for components that are not expected to show signs of deterioration can be omitted. Consequently, the processing burden of predicting the time to reach deterioration can be reduced.

[0021] Furthermore, in a diagnostic system relating to a sixth embodiment that is dependent on any one of the third to fifth embodiments, the production performance information includes mounting count information indicating the number of times the component has been mounted on the substrate by the transfer head, and the prediction unit may estimate the future deterioration state of the component based on the mounting count information and the deterioration information.

[0022] This allows for an accurate estimation of the degree of deterioration for future wear cycles based on the degree of deterioration for past wear cycles.

[0023] The embodiments will be described in detail below with reference to the drawings.

[0024] The embodiments described below are all comprehensive or specific examples. The numerical values, shapes, materials, components, arrangement and connection configurations of components, steps, and the order of steps shown in the following embodiments are examples only and are not intended to limit this disclosure. Furthermore, among the components in the following embodiments, those not described in the independent claim representing the highest-level concept will be described as optional components.

[0025] Furthermore, each figure is a schematic diagram and not necessarily a strictly accurate representation. Also, the same component is denoted by the same reference numeral in each figure.

[0026] (Embodiment) [Configuration of the component mounting device] Figure 1 is a plan view of the component mounting device in this embodiment. In other words, Figure 1 shows the internal configuration of the component mounting device as viewed from above. In this disclosure, the vertical direction is referred to as the Z-axis direction or up-down direction, one direction on a plane perpendicular to the vertical direction is referred to as the Y-axis direction or depth direction, and the direction perpendicular to the Y-axis direction on that perpendicular plane is referred to as the X-axis direction, left-right direction or lateral direction. In this disclosure, the positive side of the Z-axis direction is upward or up, and the negative side of the Z-axis direction is downward or down. In this disclosure, the positive side of the Y-axis direction is the back side or back, and the negative side of the Y-axis direction is the front side or front. In this disclosure, the positive side of the X-axis direction is the right side or right, and the negative side of the X-axis direction is the left side or left.

[0027] The component mounting device 1 in this embodiment is a production facility that produces mounted substrates by using a transfer head 8 to pick up components and mount them onto a substrate 3. This component mounting device 1 comprises a base 1a, a substrate transport mechanism 2, two component supply units 4, a Y-axis beam 6, two X-axis beams 7, two transfer heads 8, two substrate recognition cameras 12, and two component recognition cameras 11.

[0028] The base 1a is a platform for mounting the substrate transport mechanism 2, the Y-axis beam 6, two X-axis beams 7, and two component recognition cameras 11.

[0029] The substrate transport mechanism 2 is equipped with two rails aligned along the X-axis and is positioned at the center of the base 1a in the Y-axis direction. The substrate transport mechanism 2 transports the substrate 3 brought in from the upstream side (for example, the negative side in the X-axis direction) and positions and holds the substrate 3 on the mounting stage, which is the position for performing component mounting work.

[0030] The two component supply units 4 are arranged so as to sandwich the substrate transport mechanism 2 in the Y-axis direction. Multiple tape feeders 5 are arranged along the X-axis direction in the component supply units 4. The tape feeders 5 are also simply called feeders and supply components. Specifically, the tape feeders 5 supply components by pitching a carrier tape containing components in the tape feeding direction. The components are electronic components such as IC (Integrated Circuit) chips.

[0031] The Y-axis beam 6 is positioned along the Y-axis direction on the positive side in the X-axis direction (the right end in the example shown in Figure 1) of the upper surface of the base 1a.

[0032] The two X-axis beams 7 are positioned along the X-axis and are mounted on the Y-axis beam 6 so as to be movable in the Y-axis direction. For example, each of the two X-axis beams 7 moves horizontally in the Y-axis direction by being driven by the drive mechanism of the Y-axis beam 6.

[0033] Each of the two transfer heads 8 is mounted on the X-axis beam 7 via a coupling plate 8a so as to be movable in the X-axis direction. Therefore, the transfer heads 8 move in the X-axis and Y-axis directions by the Y-axis beam 6 and the X-axis beam 7. Multiple nozzle units 9 capable of picking up, holding, and raising / lowering parts are detachably mounted on the transfer heads 8. By moving in the X-axis and Y-axis directions, the transfer heads 8 pick up parts supplied from the part supply unit 4 with the nozzle units 9 and mount or attach those parts to the mounting points on the substrate 3 positioned by the substrate transport mechanism 2.

[0034] Each of the two substrate recognition cameras 12 is located on the lower side of the X-axis beam 7 and is mounted on the coupling plate 8a so as to move integrally with the transfer head 8. Specifically, the substrate recognition cameras 12 are mounted on the coupling plate 8a in a position with the imaging direction facing downwards. The substrate recognition cameras 12 move together with the transfer head 8 onto the substrate 3 which is positioned by the substrate transport mechanism 2, and image the substrate 3 in order to recognize its position and type.

[0035] The two component recognition cameras 11 are arranged on the base 1a so as to sandwich the substrate transport mechanism 2 in the Y-axis direction. Each of the two component recognition cameras 11 captures an image of the component from the negative Z-axis side when the transfer head 8, which is corresponding to that component recognition camera 11, moves over the component while holding the component. Recognition processing is performed on the image obtained from this capture to identify the position, angle, and type of the component held by the transfer head 8.

[0036] Figure 2 is a perspective view of the transfer head 8 used in the component mounting device 1.

[0037] As described above, the transfer head 8 is mounted on the X-axis beam 7 via a coupling plate 8a. Multiple nozzle units 9 are arranged side by side on the transfer head 8. Each nozzle unit 9 has a nozzle drive unit 9a, a nozzle shaft 13, a nozzle mounting unit 14, and a suction nozzle 15.

[0038] The nozzle drive unit 9a has a nozzle lifting mechanism that raises and lowers a lifting shaft using a linear motor. The nozzle shaft 13 is coupled to the lifting shaft of the nozzle drive unit 9a so as to extend downward from the nozzle drive unit 9a. The nozzle mounting unit 14 is coupled to the lower end of the nozzle shaft 13. The suction nozzle 15 is detachably mounted on the lower side of the nozzle mounting unit 14 and holds and holds parts by vacuum suction. In such a transfer head 8, the linear motor of the nozzle drive unit 9a drives each of the multiple nozzle units 9, causing the suction nozzles 15 mounted on the nozzle mounting unit 14 to move up and down individually. Each nozzle mounting unit 14 is equipped with a type of suction nozzle 15 that corresponds to the size and shape of the part to be suctioned.

[0039] Furthermore, the component mounting device 1 is equipped with an air control mechanism for drawing air into the suction nozzle 15 and blowing air out from the suction nozzle 15.

[0040] Figure 3 shows an example of the configuration of the air control mechanism in this embodiment.

[0041] The air control mechanism of the component mounting device 1 comprises a vacuum pump 19, an air supply source 21, an atmospheric supply source 22, the aforementioned multiple nozzle units 9, and a nozzle control unit 23.

[0042] The vacuum pump 19 generates negative pressure (also called a vacuum). This vacuum pump 19 is connected to the input port P1 of the switching valve 18 of the nozzle unit 9 via an air passage. The air supply source 21 is connected to the input port P3 of the blow valve 20 of the nozzle unit 9 via an air passage and supplies positive-pressure air to the blow valve 20. The atmospheric air supply source 22 is connected to the input port P4 of the blow valve 20 via an air passage and supplies air at, for example, atmospheric pressure to the blow valve 20. The atmospheric air supply source 22 can also be realized by opening the input port P4 of the blow valve 20.

[0043] The nozzle control unit 23 controls the switching valve 18 and the blow valve 20 of the nozzle unit 9.

[0044] The nozzle unit 9 includes a switching valve 18, a blow valve 20, a flow sensor 16, a nozzle shaft 13, a nozzle mounting section 14, an adsorption nozzle 15, an air tube 40, and a filter 41.

[0045] The air tube 40 is connected to the nozzle shaft 13. The cavities formed inside the air tube 40, the nozzle shaft 13, the nozzle mounting section 14, and the suction nozzle 15 are in communication with each other. Therefore, air can flow from the upper end of the air tube 40 to the lower end of the suction nozzle 15, and conversely, air can flow from the lower end of the suction nozzle 15 to the upper end of the air tube 40.

[0046] Furthermore, the filter 41 is located inside the nozzle mounting section 14 and purifies the air passing through it.

[0047] The blow valve 20 consists of a solenoid valve having two input ports P3 and P4 and an output port A2. The output port A2 of the blow valve 20 is connected to the input port P2 of the switching valve 18 via an air passage. The blow valve 20 switches the air passage by opening and closing the solenoid valve in response to control by the nozzle control unit 23. In other words, the blow valve 20 switches the air passage between a first passage and a second passage.

[0048] The first flow path is a flow path through which air flows between the air supply source 21 and the switching valve 18 via the blow valve 20. For example, positive pressure air is supplied from the air supply source 21 to the switching valve 18 along the first flow path. The second flow path is a flow path through which air flows between the atmospheric supply source 22 and the switching valve 18 via the blow valve 20. For example, atmospheric pressure air is supplied from the atmospheric supply source 22 to the switching valve 18 along the second flow path.

[0049] The switching valve 18 is composed of a solenoid valve having two input ports P1 and P2 and an output port A1. The output port A1 of the switching valve 18 is connected to the suction nozzle 15 via an output path 17, an air tube 40, a nozzle shaft 13, and a nozzle mounting section 14, which are air passages. The switching valve 18 switches the air passage by opening and closing the solenoid valve in response to control by the nozzle control unit 23. In other words, the switching valve 18 switches the air passage between a third passage and a fourth passage.

[0050] The third flow path is a passage through which air flows between the suction nozzle 15 and the vacuum pump 19 via the nozzle mounting section 14, the air tube 40, the output path 17, and the switching valve 18. When the vacuum pump 19 is driven and air flows along this third flow path, the air flows in the direction indicated by arrow b in Figure 3. Then, the surrounding air is drawn into the suction hole formed in the suction holding surface 15a at the lower end of the suction nozzle 15. As a result, the part D is attracted to and held by the suction holding surface 15a.

[0051] The fourth flow path is a flow path through which air flows between the suction nozzle 15 and the blow valve 20 via the switching valve 18, output path 17, air tube 40, nozzle shaft 13, and nozzle mounting section 14. When the air flow path is switched to the first flow path by the blow valve 20 and the air flow path is switched to the fourth flow path by the switching valve 18, air flows in the direction indicated by arrow a in Figure 3, and air is blown out from the suction holes of the suction nozzle 15. In other words, driven by the air supply source 21, air at a positive pressure flows from the air supply source 21 to the suction nozzle 15 via the blow valve 20 and the switching valve 18, and is blown out from the suction holes formed on the suction holding surface 15a of the suction nozzle 15. Furthermore, when the air flow path is switched to the second flow path by the blow valve 20 and the air flow path is switched to the fourth flow path by the switching valve 18, air flows so that the air pressure in the second and fourth flow paths becomes atmospheric pressure. In other words, the air pressure inside each of the following components becomes atmospheric pressure: the air supply source 22, the blow valve 20, the switching valve 18, the air tube 40, the nozzle shaft 13, the nozzle mounting section 14, and the suction nozzle 15.

[0052] The flow sensor 16 measures the flow rate of air flowing through the output path 17 and outputs flow rate information d1 that shows the measurement result. For example, the flow rate of air flowing in the direction indicated by arrow a in Figure 3 is measured as a positive flow rate, and the flow rate of air flowing in the direction indicated by arrow b in Figure 3 is measured as a negative flow rate. The flow rate information d1 shows, for example, the change in flow rate over time as a waveform. Such flow rate information d1 is output from each of the flow sensors 16 of the multiple nozzle units 9 included in the transfer head 8.

[0053] When the suction nozzle 15 draws in air while the part D is in contact with the suction holding surface 15a, the part D is vacuum-adsorbed by the suction nozzle 15. At this time, the air flow rate measured by the flow sensor 16 is approximately zero. When the suction nozzle 15 is drawing in ambient air while the part D is not in contact with the suction holding surface 15a, the air flow rate measured by the flow sensor 16 is negative. Also, when air is blowing out from the suction nozzle 15, the air flow rate measured by the flow sensor 16 is positive. Furthermore, when the air pressure inside the air supply source 22, blow valve 20, switching valve 18, air tube 40, nozzle shaft 13, nozzle mounting part 14, and suction nozzle 15 is atmospheric pressure, the air flow rate measured by the flow sensor 16 is approximately zero.

[0054] Figure 4 is a block diagram showing an example of the functional configuration of the component mounting device 1 in this embodiment.

[0055] As described above, the component mounting device 1 includes a substrate transport mechanism 2, a component supply unit 4, a head moving mechanism 10, a component recognition camera 11, a substrate recognition camera 12, a vacuum pump 19, an air supply source 21, an atmospheric supply source 22, and a transfer head 8. The component mounting device 1 further includes a device control unit 30, a device storage unit 31, an input unit 32, a presentation unit 33, and a diagnostic system 100.

[0056] The head movement mechanism 10 is a mechanism that includes the Y-axis beam 6 and X-axis beam 7 described above.

[0057] The device control unit 30 controls each component of the component mounting device 1. For example, the device control unit 30 is composed of a CPU (Central Processing Unit) or a processor.

[0058] The device storage unit 31 is a recording medium that stores various data necessary for the component mounting device 1 to mount components D onto the circuit board 3. For example, the device storage unit 31 may be a hard disk drive, RAM (Random Access Memory), ROM (Read Only Memory), or semiconductor memory. Such a device storage unit 31 may be volatile or non-volatile. The device storage unit 31 may also store a computer program that is read and executed by the device control unit 30. In this case, the device control unit 30 controls each component of the component mounting device 1 by reading and executing the computer program.

[0059] The input unit 32 is configured as, for example, a keyboard, touch sensor, touchpad, or mouse. Such an input unit 32 receives input operations from an operator who is producing a mounting board using the component mounting device 1, and outputs a signal corresponding to that input operation to the device control unit 30 or the diagnostic system 100, etc.

[0060] The presentation unit 33 receives a presentation signal from the device control unit 30 or the diagnostic system 100 and outputs at least one of an image and sound corresponding to the presentation signal. Specifically, the presentation unit 33 is a display such as a liquid crystal display or an organic EL (Electro-Luminescence) display. In this case, the presentation unit 33 displays an image corresponding to the presentation signal. Alternatively, the presentation unit 33 may be a speaker or the like. In this case, the presentation unit 33 outputs sound corresponding to the presentation signal. Furthermore, the presentation unit 33 may include both a display and a speaker.

[0061] The diagnostic system 100 diagnoses the deterioration state of each component included in each of the multiple nozzle units 9 of the transfer head 8. In other words, the diagnostic system 100 estimates the current degree of deterioration of each component and further estimates the future degree of deterioration. The diagnostic system 100 also predicts the maintenance timing for those components. Each component included in the nozzle unit 9 is a component through which air passes. Specifically, each component is an air tube 40, a filter 41, a valve, or a shaft. The valve is at least one of the switching valve 18 and the blow valve 20. The shaft is, for example, the nozzle shaft 13.

[0062] [Diagnostic system configuration and operation] Figure 5 is a block diagram showing an example of the functional configuration of the diagnostic system 100 in this embodiment.

[0063] The diagnostic system 100 uses flow rate information d1, production plan information d2, and production performance information d3 to diagnose the deterioration state of each component contained in each of the multiple nozzle units 9, and outputs a presentation signal indicating the diagnostic result to the presentation unit 33. As a result, the diagnostic result is presented from the presentation unit 33 in the form of an image, sound, or the like.

[0064] Such a diagnostic system 100 includes an acquisition unit 101, a maintenance processing unit 102, a prediction unit 103, an output unit 104, a learning unit 110, a model storage unit 120, and a deterioration identification unit 130.

[0065] The acquisition unit 101 acquires flow rate information d1, production plan information d2, and production results information d3. Flow rate information d1 is information regarding the flow rate of air flowing to the transfer head 8 in the component mounting device 1, which uses the transfer head 8 to pick up and mount components D onto the substrate 3. Production plan information d2 is information indicating a plan to produce mounted substrates using the component mounting device 1. In other words, production plan information d2 is information indicating a production plan to produce mounted substrates, which are substrates 3 with components D mounted on them, using production equipment. For example, for each of the multiple nozzle units 9 of the component mounting device 1, production plan information d2 shows the number of times (hereinafter referred to as the number of mountings) in which components D are picked up and mounted onto the substrate 3 by that nozzle unit 9 as described above. More specifically, production plan information d2 shows the number of mountings at each of several future points in time. Production results information d3 is information indicating the actual operation of the component mounting device 1 to produce mounted substrates in accordance with the production plan information d2. In other words, production performance information d3 is information that shows the actual number of times mounted substrates have been produced using the production equipment. For example, for each of the multiple nozzle units 9 of the component mounting device 1, production performance information d3 shows the number of times up to the present time that a component D has been attracted and mounted onto the substrate 3 by that nozzle unit 9. More specifically, production performance information d3 shows the number of times mounted at each of several points in the past.

[0066] The acquisition unit 101 outputs flow rate information d1 to the learning unit 110 and the deterioration identification unit 130. Furthermore, the acquisition unit 101 outputs production plan information d2 and production performance information d3 to the prediction unit 103.

[0067] The learning unit 110 acquires flow rate information d1 from the acquisition unit 101 as training data, generates a diagnostic model using machine learning based on the flow rate information d1, and stores the diagnostic model in the model storage unit 120. The diagnostic model is a machine learning model used to diagnose the deterioration state of each component included in the nozzle unit 9. In a specific example, the diagnostic model is a neural network. Such a diagnostic model is generated by the learning unit 110 performing machine learning so that, in response to the input of flow rate information d1, it outputs information indicating the deterioration state of each component included in the nozzle unit 9 corresponding to the flow rate information d1. In other words, the diagnostic model can be said to show the relationship between the flow rate information d1 and the deterioration state of each component included in the nozzle unit 9.

[0068] The model storage unit 120 is a recording medium for storing diagnostic models. Such a model storage unit 120 may be a hard disk drive, RAM, ROM, or semiconductor memory.

[0069] The deterioration identification unit 130 identifies the deterioration state of each of the multiple components included in the nozzle unit 9 corresponding to the flow rate information d1 of the transfer head 8, based on the flow rate information d1. Specifically, the deterioration identification unit 130 acquires the flow rate information d1 from the acquisition unit 101 and acquires a diagnostic model from the model storage unit 120. Then, the deterioration identification unit 130 inputs the flow rate information d1 into the diagnostic model and acquires the deterioration information output from the diagnostic model. This deterioration information indicates the deterioration state of each component included in the nozzle unit 9. By acquiring such deterioration information, the deterioration identification unit 130 identifies the deterioration state of each component. In other words, the deterioration identification unit 130 identifies the deterioration state by using the diagnostic model. The deterioration identification unit 130 outputs the deterioration information indicating the identified deterioration state to the output unit 104, the maintenance processing unit 102, and the prediction unit 103. The deterioration information is information indicating the past or present deterioration state of each component included in the production equipment, which is the parts mounting device 1. Furthermore, the deterioration information may indicate the deterioration state of each component at multiple past points in time.

[0070] The maintenance processing unit 102 acquires deterioration information from the deterioration identification unit 130 and generates maintenance information regarding the maintenance of the components included in the nozzle unit 9 based on that deterioration information. The maintenance processing unit 102 then outputs this maintenance information to the output unit 104.

[0071] The prediction unit 103 estimates the future deterioration state of each component included in the nozzle unit 9 of the transfer head 8. Furthermore, the prediction unit 103 predicts the time when the deterioration state will reach a predetermined deterioration state as the maintenance time. In other words, the prediction unit 103 acquires information indicating the deterioration state identified by the deterioration identification unit 130 as deterioration information. Then, based on this deterioration information and the production plan information d2 and production performance information d3 acquired by the acquisition unit 101, the prediction unit 103 predicts the time when the future deterioration state of each component will reach a predetermined deterioration state. The prediction unit 103 outputs the predicted arrival time information to the output unit 104.

[0072] The output unit 104 outputs degradation information indicating the identified degradation state. That is, when the output unit 104 acquires degradation information from the degradation identification unit 130, it outputs that degradation information as a presentation signal to the presentation unit 33. As a result, the presentation unit 33 presents the content of the degradation information (i.e., the degradation state). The output unit 104 also outputs arrival time information indicating the predicted arrival time. That is, when the output unit 104 acquires arrival time information from the prediction unit 103, it outputs that arrival time information as a presentation signal to the presentation unit 33. As a result, the presentation unit 33 presents the content of that arrival time information (i.e., the arrival time). Furthermore, the output unit 104 outputs maintenance information. That is, when the output unit 104 acquires maintenance information from the maintenance processing unit 102, it outputs that maintenance information as a presentation signal to the presentation unit 33. As a result, the presentation unit 33 presents the content of that maintenance information (i.e., content related to maintenance).

[0073] Figure 6 is a diagram illustrating the learning unit 110 and the diagnostic model.

[0074] The learning unit 110 comprises a feature extraction unit 111 and a learning processing unit 112. When the feature extraction unit 111 acquires flow rate information d1, which is training data, from the acquisition unit 101, it extracts multiple types of features related to the air flow rate from the flow rate information d1 and outputs training feature data da1, which shows these multiple types of features, to the learning processing unit 112.

[0075] The flow rate information d1 is, for example, a flow rate waveform showing the change in air flow rate over time as measured by the flow rate sensor 16. This flow rate waveform may be obtained, for example, by having the nozzle unit 9 repeatedly and intermittently suck or blow air under control by the nozzle control unit 23 when the component mounting device 1 is not producing mounting substrates. Alternatively, the flow rate waveform may be obtained when the suction nozzle 15 is not mounted on the nozzle mounting section 14. The feature extraction unit 111 extracts characteristic times or flow rates indicated by the flow rate waveform as features. Specifically, the features are numerical values ​​such as the positive peak flow rate, the negative peak flow rate, the steady flow rate, the response time, or the steady time. Alternatively, the features may be values ​​calculated by an operation using two or more of these numerical values, or they may be vectors consisting of these numerical values. The response time is, for example, the time from when the air flow path is switched by the switching valve 18 and the blow valve 20 until the air flow rate stabilizes. The steady time is the time during which the stable flow rate continues. Steady-state flow rate is the stable flow rate.

[0076] The learning processing unit 112 obtains training feature data da1 from the feature extraction unit 111, generates a diagnostic model 121 by performing machine learning using the feature data da1, and stores the diagnostic model 121 in the model storage unit 120. The machine learning is, for example, learning using a neural network or a deep neural network. Furthermore, the machine learning may be supervised learning or unsupervised learning. In addition, the machine learning may be random forest, SVM (Support Vector Machine), Gaussian process regression, SVR (Support Vector Regression), or random forest regression. The machine learning may also be learning that generates an exponential model, power model, logarithmic model, Gompertz model, or Lloyd-Lipow model as the diagnostic model 121.

[0077] Such a learning processing unit 112 comprises an anomaly learning unit 112a, a classification learning unit 112b, and a degradation degree learning unit 112c. The anomaly learning unit 112a generates an anomaly judgment model 121a by machine learning using training feature data da1. This anomaly judgment model 121a is, for example, a model that, given the input of feature data da1, outputs anomaly judgment result information indicating whether or not the nozzle unit 9 corresponding to the feature data da1 is abnormal. In the case of supervised learning, the anomaly learning unit 112a learns using training feature data da1 and training data corresponding to at least one feature shown in the feature data da1. The training data indicates whether or not the nozzle unit 9 is abnormal. The anomaly judgment model 121a, for example, indicates regions where the nozzle unit 9 is abnormal and regions where the nozzle unit 9 is normal in a two-dimensional space with each of the two types of features as coordinate axes.

[0078] The classification learning unit 112b generates an anomaly classification model 121b by machine learning using the training feature data da1. This anomaly classification model 121b is a model that, given at least the input feature data da1, outputs anomaly classification information indicating which component of the nozzle unit 9 corresponding to the feature data da1 is anomaly. In the case of supervised learning, the classification learning unit 112b performs learning using the training feature data da1 and training data corresponding to at least one feature shown in the feature data da1. The training data indicates which component of the nozzle unit 9 is anomaly.

[0079] The degradation degree learning unit 112c generates a degradation degree estimation model 121c by machine learning using the training feature data da1. This degradation degree estimation model 121c is a model that, given at least the feature data da1 as input, outputs degradation degree information indicating the degree of degradation of the components of the nozzle unit 9 corresponding to the feature data da1. In the case of supervised learning, the degradation degree learning unit 112c learns using the training feature data da1 and training data corresponding to at least one feature shown in the feature data da1. The training data indicates the degree of degradation of the components of the nozzle unit 9.

[0080] Figure 7 is a diagram illustrating the deterioration identification section 130 and the diagnostic model.

[0081] The degradation identification unit 130 comprises a feature extraction unit 131 and a identification processing unit 132. When the feature extraction unit 131 acquires flow rate information d1 from the acquisition unit 101, it extracts multiple types of features related to the air flow rate from the flow rate information d1 and outputs feature data da1 representing these multiple types of features to the identification processing unit 132. In other words, the feature extraction unit 131 of the degradation identification unit 130 has the same configuration and function as the feature extraction unit 111 of the learning unit 110.

[0082] The identification processing unit 132 acquires feature data da1 from the feature extraction unit 131 and the diagnostic model 121 from the model storage unit 120. The identification processing unit 132 then inputs the feature data da1 into the diagnostic model 121 and acquires the degradation information db output from the diagnostic model 121. This identifies the degradation state of each component included in the nozzle unit 9 corresponding to the feature data da1. Specifically, the identification processing unit 132 comprises an anomaly processing unit 132a, a classification processing unit 132b, and a degradation degree processing unit 132c.

[0083] The anomaly processing unit 132a obtains the anomaly judgment model 121a from the model storage unit 120, and inputs the feature data da1 into the anomaly judgment model 121a. As a result, the anomaly processing unit 132a obtains the anomaly judgment result information db1 output from the anomaly judgment model 121a. This anomaly judgment result information db1 indicates whether the nozzle unit 9 corresponding to the feature data da1 is abnormal or not. In other words, the anomaly processing unit 132a determines whether the nozzle unit 9 corresponding to the feature data da1 is abnormal or not.

[0084] The classification processing unit 132b obtains the anomaly classification model 121b from the model storage unit 120, and inputs the feature data da1 into the anomaly classification model 121b. As a result, the classification processing unit 132b obtains the anomaly classification information db2 output from the anomaly classification model 121b. This anomaly classification information db2 indicates which component of the nozzle unit 9 corresponding to the feature data da1 is abnormal. In other words, the classification processing unit 132b classifies the state of each of the multiple components included in the nozzle unit 9 corresponding to the feature data da1 as either normal or abnormal.

[0085] The degradation degree processing unit 132c acquires the degradation degree estimation model 121c from the model storage unit 120, and inputs the feature data da1 into the degradation degree estimation model 121c. This allows the degradation degree processing unit 132c to acquire the degradation degree information db3 output from the degradation degree estimation model 121c. This degradation degree information db3 indicates the degree of degradation of the components of the nozzle unit 9. In other words, the degradation degree processing unit 132c estimates the degree of degradation of the components of the nozzle unit 9 using the degradation degree estimation model 121c. Thus, the degradation identification unit 130 identifies the degradation state by estimating the degree of degradation of each component of at least one component included in the nozzle unit 9. Furthermore, the higher the degree of degradation, the greater the value.

[0086] Figure 8 shows a specific example of a degradation information database.

[0087] The deterioration identification unit 130 generates and outputs deterioration information db for each of the multiple nozzle units 9 included in the transfer head 8. As shown in Figure 8, the deterioration information db includes abnormality judgment result information db1, abnormality classification information db2, and deterioration degree information db3. In a specific example, abnormality judgment result information db1 indicates that the nozzle unit 9 corresponding to that abnormality judgment result information db1 is abnormal. Then, abnormality classification information db2 indicates whether each component included in the nozzle unit 9 is normal or abnormal. For example, abnormality classification information db2 indicates that the filter 41 is normal, the valve such as the switching valve 18 is normal, the air tube 40 is abnormal, and the shaft which is the nozzle shaft 13 is normal. Deterioration degree information db3 indicates the degree of deterioration of each component included in the nozzle unit 9 by an integer value from 0 to 10. The closer the deterioration degree is to 10, the greater the degree of deterioration, and the closer the deterioration degree is to 0, the less the degree of deterioration. For example, the degradation level information db3 shows that the degradation level of the filter 41 is "1", the degradation level of the valve is "5", the degradation level of the air tube 40 is "9", and the degradation level of the shaft is "6".

[0088] In this embodiment, the contents of such deterioration information database are presented by the presentation unit 33. Therefore, an operator performing work using the parts mounting device 1 can understand the deterioration status of at least one component included in the transfer head 8 from the output deterioration information database. As a result, it is possible to diagnose the detailed state of the transfer head 8 rather than a simple state such as whether the component is normal or not. Since the deterioration status of each of multiple components is diagnosed, the operator can easily determine whether maintenance is necessary for those components or whether preparations for such maintenance should be made. Therefore, the operator can avoid performing unnecessary maintenance on components that do not require maintenance. Alternatively, the operator can avoid performing unnecessary maintenance or preparations for maintenance on components that do not require maintenance. As a result, maintenance efficiency can be improved, maintenance time can be shortened, and maintenance costs can be reduced. In this way, maintenance by operators of the parts mounting device 1, which is a production facility, can be more appropriately supported.

[0089] The maintenance processing unit 102 generates maintenance information for the components included in the nozzle unit 9 corresponding to the degradation information database, based on such degradation information database. Specifically, for each of the multiple components included in the nozzle unit 9, the maintenance processing unit 102 determines whether the estimated degree of degradation for that component exceeds a threshold, and generates maintenance information for the component whose degree of degradation exceeds the threshold. This maintenance information may include, for example, maintenance alarm information prompting the implementation of maintenance on the component, or maintenance forecast information prompting preparation for implementation of maintenance on the component. Maintenance alarm information can be described as an alarm or warning to inform the worker that maintenance is needed immediately. Maintenance forecast information can be described as a forecast or notice to inform the worker in advance that the time when maintenance is needed is approaching. When the maintenance processing unit 102 generates maintenance information such as maintenance alarm information and maintenance forecast information, it outputs the maintenance information to the output unit 104.

[0090] When the output unit 104 receives maintenance information from the maintenance processing unit 102, it outputs that maintenance information as a presentation signal to the presentation unit 33. As a result, the content of the maintenance information is presented to the presentation unit 33 in the form of an image or sound.

[0091] Furthermore, the deterioration identification unit 130 may determine whether there is an abnormality in each of the one or more components included in the nozzle unit 9 of the transfer head 8, and estimate the degree of deterioration of each of the at least one component that is determined to be abnormal. In other words, the deterioration identification unit 130 may determine whether there is an abnormality in a component, and if it determines that there is an abnormality, it may estimate the degree of deterioration of that component.

[0092] Figure 9 is a diagram illustrating the input and output of the error processing unit 132a, the classification processing unit 132b, and the degradation degree processing unit 132c.

[0093] The anomaly processing unit 132a, the classification processing unit 132b, and the degradation degree processing unit 132c may each add new information to the input data before outputting it. In other words, the anomaly processing unit 132a acquires the feature data da1 output from the feature extraction unit 131 as input data, adds new information, namely the anomaly judgment result information db1, to that input data, and outputs it.

[0094] The classification processing unit 132b acquires the feature data da1 and anomaly judgment result information db1 output from the anomaly processing unit 132a as input data, adds new information, anomaly classification information db2, to the input data, and outputs it.

[0095] The degradation degree processing unit 132c acquires the feature data da1, anomaly judgment result information db1, and anomaly classification information db2 output from the classification processing unit 132b as input data, and adds new information, degradation degree information db3, to the input data before outputting it. The degradation degree information db3 output at this time may indicate the degree of degradation of the component members (also called anomaly members) that are classified as anomaly in the anomaly classification information db2. Furthermore, the degradation degree processing unit 132c may acquire the feature data da1 output from the feature extraction unit 131 as input data. In this case, the degradation degree processing unit 132c may output degradation degree information db3 that indicates the degree of degradation of the component members (also called normal members) that are classified normally in the anomaly classification information db2, based on the feature data da1.

[0096] Figure 10 is a diagram illustrating the processing of the prediction unit 103.

[0097] The prediction unit 103 acquires production performance information d3 from the acquisition unit 101 and deterioration information db from the deterioration identification unit 130. The production performance information d3 includes mounting count information for each of the multiple nozzle units 9 of the transfer head 8, indicating the number of times that the component D has been mounted on the substrate 3 by that nozzle unit 9. Based on the mounting count information and the deterioration information db, the prediction unit 103 estimates the future deterioration state of each component included in each of the multiple nozzle units 9.

[0098] For example, the graph in Figure 10 shows the relationship between the degree of degradation of a component included in one nozzle unit 9 and the number of times that component D is mounted onto the substrate 3 by that nozzle unit 9. The horizontal axis of the graph represents the number of mountings, and the vertical axis represents the degree of degradation.

[0099] The prediction unit 103 identifies degradation points (black circles in Figure 10) that indicate the current or past number of installations and the degree of degradation at that installation count, based on the production performance information d3 and the degradation information db. Specifically, for each current or past point in time, the prediction unit 103 identifies degradation points at each current or past point in time by associating the number of installations at that time, as shown in the installation count information of the production performance information d3, with the degree of degradation shown in the degradation information db obtained at that time.

[0100] When the prediction unit 103 identifies one or more deterioration points, it calculates an approximate curve for those deterioration points as a deterioration approximation curve. This deterioration approximation curve may be calculated, for example, by the least squares method. In other words, the prediction unit 103 estimates the future deterioration state by calculating this deterioration approximation curve. Specifically, the prediction unit 103 estimates that the relationship between the number of installations greater than the current number of installations m0 and the degree of deterioration of the component is shown by the deterioration approximation curve. Therefore, the prediction unit 103 determines that the predicted deterioration points (square marks in Figure 10) that indicate the degree of deterioration at the number of installations greater than the number of installations m lie on this deterioration approximation curve. In this way, the prediction unit 103 estimates the future deterioration state of the component from the present time onward, with the future degree of deterioration being greater as the degree of deterioration of the component increases.

[0101] Next, the prediction unit 103 predicts the time at which the future deterioration state of the component will reach a predetermined deterioration state. This predetermined deterioration state is either a first predetermined deterioration state requiring maintenance of the component, or a second predetermined deterioration state requiring preparation for such maintenance. In other words, the first predetermined deterioration state is the deterioration state of the component when the aforementioned maintenance alarm information is generated for the component and the content of that maintenance alarm information is presented by the presentation unit 33. The second predetermined deterioration state is the deterioration state of the component when the aforementioned maintenance forecast information is generated for the component and the content of that maintenance forecast information is presented by the presentation unit 33.

[0102] Specifically, the prediction unit 103 predicts, based on the production plan information d2, the time at which the future degree of deterioration will reach a first threshold corresponding to a first defined deterioration state, or a second threshold corresponding to a second defined deterioration state. In other words, the prediction unit 103 identifies the number of installations m1 when the degree of deterioration is at the first threshold in the deterioration approximation curve. Then, the prediction unit 103 refers to the production plan information d2 and identifies the period T1 from the current number of installations m0 to the number of installations m1 from that production plan information d2. The production plan information d2 shows the relationship between the time elapsed for each nozzle unit 9 and the number of installations. The prediction unit 103 predicts the timing after the identified period T1 has elapsed from the present as the time at which the future deterioration state of the component will reach the first defined deterioration state. This time at which it will reach is also called the maintenance implementation time.

[0103] Similarly, the prediction unit 103 identifies the number of installations m2 when the degree of deterioration reaches the second threshold in the deterioration approximation curve. The second threshold is a smaller value than the first threshold. The prediction unit 103 then refers to the production plan information d2 and identifies the period T2 from the current number of installations m0 to the number of installations m2. The prediction unit 103 predicts the timing after the identified period T2 has elapsed from the current time as the arrival time when the future deterioration state of the component reaches the second specified deterioration state. This arrival time is also called the maintenance preparation time. The above-mentioned maintenance implementation time and maintenance preparation time are collectively referred to as the maintenance time. The arrival time information indicates that maintenance time.

[0104] In this embodiment, the time when maintenance of deteriorated components is required, or the time when preparation for such maintenance is required, is predicted as the arrival time, and arrival time information indicating that arrival time is output. Therefore, workers performing work using the component mounting device 1, which is a production facility, can grasp the arrival time, i.e., the maintenance implementation time or the maintenance preparation time. As a result, workers can perform their work in anticipation of these times, thereby improving work efficiency. Furthermore, the efficiency of maintenance can be improved. In this way, it is possible to more appropriately support workers in performing maintenance on production facilities.

[0105] [Process flow of the diagnostic system] Figure 11 is a flowchart showing an example of the processing operation related to the identification of the deterioration state by the diagnostic system 100.

[0106] First, the acquisition unit 101 of the diagnostic system 100 acquires flow rate information d1 (step S1). Next, the feature extraction unit 131 of the deterioration identification unit 130 generates feature data da1 by extracting features from the flow rate information d1 (step S2).

[0107] Then, the identification processing unit 132 of the deterioration identification unit 130 executes the processes of steps S3, S4, and S5. Specifically, the abnormality processing unit 132a of the identification processing unit 132 determines whether or not there is an abnormality in the nozzle unit 9 (step S3). The classification processing unit 132b of the identification processing unit 132 classifies the state of each component included in the nozzle unit 9 into abnormal and normal (step S4). The deterioration degree processing unit 132c of the identification processing unit 132 estimates the degree of deterioration of each component included in the nozzle unit 9 (step S5).

[0108] The output unit 104 outputs a degradation information database, which shows the results of the processing in steps S3, S4, and S5 performed by the identification processing unit 132 of the degradation identification unit 130, to the presentation unit 33 as a diagnostic result (step S6). As a result, the contents of the degradation information database are presented by the presentation unit 33.

[0109] Figure 12 is a flowchart showing an example of the processing operation related to maintenance instructions by the diagnostic system 100.

[0110] First, the diagnostic system 100 performs a diagnostic process (step S10). This diagnostic process is the process described in steps S1 to S5 of the flowchart in Figure 11.

[0111] Then, the maintenance processing unit 102 of the diagnostic system 100 determines whether the degree of deterioration of each component obtained in the diagnostic process of step S10 is greater than a first threshold (step S11). If the maintenance processing unit 102 determines that the degree of deterioration of a component is greater than the first threshold (Yes in step S11), it instructs the operator to perform maintenance via the output unit 104 and the presentation unit 33 (step S12). In other words, for each of the multiple components included in the nozzle unit 9, the maintenance processing unit 102 determines whether the estimated degree of deterioration for that component exceeds a first threshold. The maintenance processing unit 102 then generates maintenance alarm information prompting the operator to perform maintenance on the component whose degree of deterioration is determined to exceed the first threshold. Subsequently, the maintenance processing unit 102 outputs the maintenance alarm information to the output unit 104, and the output unit 104 outputs the maintenance alarm information to the presentation unit 33. As a result, the content of the maintenance alarm information is presented from the presentation unit 33.

[0112] In this embodiment, for example, when a component is severely deteriorated, maintenance alarm information is generated and output to prompt the performment of maintenance on that component. As a result, the maintenance alarm information is notified to the worker, and the worker can perform maintenance on the component quickly and appropriately.

[0113] On the other hand, in step S11, if the maintenance processing unit 102 determines that the degree of deterioration of a component is below the first threshold (No in step S11), it further determines whether that degree of deterioration is greater than the second threshold (step S13). Here, if the maintenance processing unit 102 determines that the degree of deterioration of a component is greater than the second threshold (Yes in step S13), it instructs the operator to prepare for maintenance via the output unit 104 and the presentation unit 33 (step S14). This second threshold is a value less than the first threshold. In other words, for each of the multiple components included in the nozzle unit 9, the maintenance processing unit 102 determines whether the estimated degree of deterioration for that component is below the first threshold and whether that degree of deterioration exceeds the second threshold, which is smaller than the first threshold. The maintenance processing unit 102 then generates maintenance forecast information that prompts preparation for maintenance of the component having a degree of deterioration that is determined to be below the first threshold but exceeds the second threshold. Subsequently, the maintenance processing unit 102 outputs the maintenance forecast information to the output unit 104, and the output unit 104 outputs the maintenance forecast information to the presentation unit 33. As a result, the content of the maintenance forecast information is presented by the presentation unit 33.

[0114] Thus, in this embodiment, for example, even if a component is not severely deteriorated, if it is approaching a state of severe deterioration, maintenance forecast information is generated and output that prompts preparation for maintenance of that component. As a result, the maintenance forecast information is notified to the worker, and the worker can prepare for the maintenance of that component in advance and appropriately.

[0115] Furthermore, the output unit 104 outputs the degradation information db, which shows the result of the diagnostic process in step S10, to the presentation unit 33 as the diagnostic result (step S6). As a result, the contents of the degradation information db are presented by the presentation unit 33.

[0116] Figure 13 is a flowchart showing an example of the processing operation related to maintenance timing prediction by the diagnostic system 100.

[0117] First, the diagnostic system 100 performs a diagnostic process (step S10). This diagnostic process is the process described in steps S1 to S5 of the flowchart in Figure 11.

[0118] Then, the acquisition unit 101 of the diagnostic system 100 acquires production plan information d2 (step S21) and production performance information d3 (step S22). Based on the acquired production performance information d3, the prediction unit 103 estimates the future degree of deterioration of each component included in the nozzle unit 9 (step S23). This future degree of deterioration is estimated based on the number of times the nozzle unit 9 will be used to install future parts D. Then, based on the future degree of deterioration and the production plan information d2 acquired in step S21, the prediction unit 103 predicts the maintenance timing for each component of the nozzle unit 9 (step S24). This maintenance timing is the maintenance implementation timing when the future degree of deterioration reaches a first threshold corresponding to a first defined deterioration state, or the maintenance preparation timing when the future degree of deterioration reaches a second threshold corresponding to a second defined deterioration state.

[0119] Next, the output unit 104 outputs arrival time information, which indicates the maintenance time predicted in step S24, to the presentation unit 33 (step S25). As a result, the content of the arrival time information (i.e., the maintenance time) is presented by the presentation unit 33.

[0120] As described above, in the diagnostic system 100 of this embodiment, the deterioration state of at least one component included in the transfer head 8 is identified, and deterioration information db indicating that deterioration state is output. As a result, maintenance by operators of the parts mounting device 1, which is a production facility, can be more appropriately supported.

[0121] Furthermore, the deterioration identification unit 130 identifies the deterioration state of a component by estimating the degree of deterioration of that component. This allows the worker to understand the deterioration state in more detail.

[0122] Furthermore, the deterioration identification unit 130 may determine whether there is an abnormality in each of the one or more components included in the transfer head 8, and estimate the degree of deterioration for at least one of the components determined to be abnormal. As a result, the degree of deterioration is estimated for the components determined to be abnormal, so the degree of deterioration can be estimated for components that are assumed to be relatively likely to be deteriorated, and the estimation of the degree of deterioration can be omitted for components that are assumed to be relatively unlikely to be deteriorated. As a result, the processing burden of estimating the degree of deterioration can be reduced. In addition, since the degree of deterioration is estimated for the components determined to be abnormal, the operator can easily understand what kind of maintenance is required or not required for the component determined to be abnormal. For example, the operator can easily understand the type of maintenance required, such as monitoring the component over time, cleaning the component, repairing the component, replacing parts included in the component, or replacing the component itself.

[0123] Furthermore, the maintenance processing unit 102 determines whether the estimated degree of deterioration for at least one component exceeds a threshold, and generates maintenance information for the component whose degree of deterioration exceeds the threshold. The output unit 104 then outputs this maintenance information. As a result, for example, maintenance information for a deteriorated component is generated and output. Consequently, the maintenance information is notified to the worker, allowing the worker to perform maintenance on the component quickly and appropriately.

[0124] Furthermore, the deterioration identification unit 130 identifies the deterioration state by using a diagnostic model 121 that shows the relationship between the flow rate information d1 and the deterioration state of at least one component. This allows for the appropriate identification of the deterioration state.

[0125] Furthermore, the diagnostic model 121 is generated by performing machine learning so that, in response to the input of flow rate information d1, it outputs information indicating the deterioration state of at least one component. This allows for improved accuracy in identifying deterioration states by appropriately performing machine learning.

[0126] Furthermore, in the diagnostic system 100 of this embodiment, the time when maintenance of deteriorated components will be required, or the time when preparations for such maintenance will be required, is predicted as the arrival time, and arrival time information indicating that arrival time is output. Therefore, it is possible to more appropriately support workers in maintaining production equipment.

[0127] Furthermore, the degradation information database shows the degradation state of the components at multiple past points in time. This allows for accurate estimation of future degradation states based on changes in the degradation state over time, thereby improving the accuracy of predicting when degradation will occur.

[0128] Furthermore, in the diagnostic system 100, the deterioration state of the components included in the transfer head 8 is identified based on the flow rate information d1, and this identified deterioration state is used to predict the arrival time. Therefore, the deterioration state can be appropriately identified, and the accuracy of the arrival time prediction can be improved.

[0129] Furthermore, the diagnostic system 100 allows for a more detailed estimation of the deterioration state as a degree of deterioration, thereby further improving the accuracy of predicting the time of onset.

[0130] Furthermore, in the diagnostic system 100, if the degree of deterioration is estimated for a component determined to be abnormal, the system can predict the time to reach deterioration for components that are expected to show signs of deterioration, while omitting the prediction of the time to reach deterioration for components that are not expected to show signs of deterioration. As a result, the processing burden of predicting the time to reach deterioration can be reduced.

[0131] Furthermore, the production performance information d3 includes mounting count information indicating the number of times component D has been mounted on the substrate 3 by the transfer head 8, and the prediction unit 103 estimates the future deterioration state of the component based on the mounting count information and the deterioration information db. This makes it possible to appropriately estimate the degree of deterioration for future mounting counts based on the degree of deterioration for past mounting counts.

[0132] (Other forms) The diagnostic system and diagnostic method relating to this disclosure have been described above based on the embodiments described above, but this disclosure is not limited to these embodiments. Various modifications to the above embodiments that a person skilled in the art could conceive of may also be included in the scope of this disclosure, as long as they do not deviate from the spirit of this disclosure.

[0133] For example, in the above embodiment, the diagnostic system 100 is located inside the component mounting device 1, but it may also be located outside the component mounting device 1, or it may be configured as, for example, a personal computer connected to the component mounting device 1.

[0134] For example, in the above embodiment, the transfer head 8 is equipped with multiple nozzle units 9, but the number of nozzle units 9 equipped in the transfer head 8 may be just one. Also, in the above embodiment, the deterioration state is identified and the maintenance timing is predicted for each of the multiple components included in the nozzle unit 9, such as the air tube 40, filter 41, and valve, but the degree of deterioration identification and maintenance timing prediction may be performed for only one component. In other words, each of the processes such as identification and prediction may be performed for at least one component included in the transfer head 8.

[0135] Furthermore, in the above embodiment, the learning unit 110 includes a feature extraction unit 111, and the degradation identification unit 130 includes a feature extraction unit 131, but the feature extraction units 111 and 131 are not required. In other words, in the above embodiment, various features are extracted from the flow rate information d1, and feature data da1 representing these features is used by the learning processing unit 112 and the identification processing unit 132. However, the flow rate information d1 may be used directly by the learning processing unit 112 and the identification processing unit 132. That is, the flow rate waveform shown in the flow rate information d1 may be used directly for machine learning, or it may be used directly for identifying the degradation state, etc.

[0136] Furthermore, in the above embodiment, the deterioration identification unit 130 is equipped with an abnormality processing unit 132a and a classification processing unit 132b, but these components do not necessarily have to be provided. In other words, if the degree of deterioration is estimated for at least one component of the transfer head 8, it is not necessary to determine whether the nozzle unit 9 is abnormal or not, and to classify those components as abnormal or normal. Also, the abnormalities determined by the abnormality processing unit 132a and the abnormalities classified by the classification processing unit 132b may be abnormalities caused by deterioration of the components, or they may be abnormalities based on factors other than deterioration.

[0137] Furthermore, in the above embodiment, the parts mounting device 1 is an example of production equipment, but the production equipment may be a different device from the parts mounting device 1. Also, the production equipment may be called work equipment. Moreover, in the above embodiment, various information is presented to the worker who performs production using the production equipment, and the worker performs maintenance on the production equipment, but this information may also be presented to a person other than the worker, and the maintenance may be performed by a person other than the worker.

[0138] Furthermore, in the above embodiment, the diagnostic system 100 includes a maintenance processing unit 102 and a prediction unit 103, but it may also include only one of these. Also, the first threshold and second threshold used in the maintenance processing unit 102 shown in Figure 12 and the first threshold and second threshold used in the prediction unit 103 shown in Figure 10 may be the same or different.

[0139] Furthermore, in the above embodiment, the diagnostic model 121 is a machine learning model, but it may also be a table that shows the relationship between various features and the state of the nozzle unit 9. The state of the nozzle unit 9 may be normal or abnormal for the nozzle unit 9, normal or abnormal for each component included in the nozzle unit 9, or the degree of deterioration of each component.

[0140] In the above embodiment, each component may be implemented by dedicated hardware or by executing a software program suitable for each component. Each component may also be implemented by a program execution unit such as a CPU (Central Processing Unit) or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory. Here, the software that implements the diagnostic system 100 of the above embodiment is a program that causes a computer to execute each step included in the flowchart shown in Figures 11 to 13.

[0141] The following cases are also included in this disclosure.

[0142] (1) Specifically, each of the above devices is a computer system consisting of a microprocessor, ROM, RAM, hard disk unit, display unit, keyboard, mouse, etc. A computer program is stored in the RAM or hard disk unit. Each device achieves its function by operating the microprocessor in accordance with the computer program. Here, a computer program is composed of a combination of multiple instruction codes that indicate commands to the computer in order to achieve a predetermined function.

[0143] (2) Some or all of the components constituting each of the above devices may be made up of a single system LSI (Large Scale Integration). The system LSI is a multi-functional LSI manufactured by integrating multiple components onto a single chip, and specifically, it is a computer system that includes a microprocessor, ROM, RAM, etc. A computer program is stored in the RAM. The system LSI achieves its function by operating the microprocessor in accordance with the computer program.

[0144] (3) Some or all of the components constituting each of the above devices may consist of a removable IC card or a standalone module. The IC card or module is a computer system consisting of a microprocessor, ROM, RAM, etc. The IC card or module may include the above-mentioned multi-functional LSI. The microprocessor operates according to a computer program, thereby enabling the IC card or module to perform its function. The IC card or module may be tamper-resistant.

[0145] (4) The disclosure may also be the methods described above. Alternatively, it may be a computer program that implements these methods using a computer, or a digital signal consisting of the computer program.

[0146] Furthermore, this disclosure may also refer to the computer program or the digital signal recorded on a computer-readable recording medium, such as a flexible disk, hard disk, CD-ROM, MO, DVD, DVD-ROM, DVD-RAM, BD (Blu-ray® Disc), semiconductor memory, etc. Alternatively, it may refer to the digital signal recorded on such a recording medium.

[0147] Furthermore, this disclosure may also describe transmitting the computer program or digital signal via telecommunications lines, wireless or wired communication lines, networks such as the Internet, data broadcasting, etc.

[0148] Furthermore, the present disclosure may also provide a computer system comprising a microprocessor and memory, wherein the memory stores the computer program, and the microprocessor operates in accordance with the computer program.

[0149] Furthermore, the program or digital signal may be implemented by another independent computer system by recording and transferring it on the recording medium, or by transferring the program or digital signal via the network or the like.

[0150] (5) The above embodiments and other forms may be combined. [Industrial applicability]

[0151] This disclosure can be used, for example, in systems for managing production equipment. [Explanation of Symbols]

[0152] 1. Component mounting device 1a Base 2. Substrate transport mechanism 3 circuit boards 4. Parts Supply Department 5 Tape feeder 6 Y-axis beam 7 X-axis beam 8 Transfer head 8a Bonding plate 9 Nozzle Unit 9a Nozzle drive unit 10 Head movement mechanism 11. Part Recognition Camera 12. Circuit board recognition camera 13 Nozzle shaft 14 Nozzle mounting section 15 Suction nozzle 15a Suction holding surface 16 Flow Sensor 17 Output path 18. Switching valve 19 Vacuum pump 20 Blow valve 21 Air supply source 22. Atmospheric sources 23 Nozzle control unit 30 Device Control Unit 31 Device storage 32 Input section 33 Presentation section 40 Air Tubes 41 Filters 100 diagnostic systems 101 Acquisition Department 102 Maintenance Processing Unit 103 Prediction Section 104 Output section 110 Learning Department 111 Feature Extraction Unit 112 Learning Processing Unit 112a Abnormal Learning Unit 112b Classification and Learning Department 112c Degradation Degree Learning Unit 120 Model Storage Unit 121 Diagnostic Models 121a Anomaly detection model 121b Anomaly Classification Model 121c Degradation Estimation Model 130 Deterioration identification section 131 Feature Extraction Unit 132 Specific Processing Unit 132a Error Processing Unit 132b Classification Processing Unit 132c Degradation treatment section d1 Flow rate information d2 Production Plan Information d3 Production Performance Information da1 Feature Data DB degradation information db1 abnormality judgment result information db2 anomaly classification information DB3 degradation level information

Claims

1. (i) Production plan information indicating a production plan for producing a mounting board, which is a substrate on which components are mounted, using production equipment, and (ii) Production performance information indicating the actual production of the mounting board using the production equipment, A prediction unit predicts the time at which the future deterioration state of the components included in the production equipment will reach a predetermined deterioration state, based on deterioration information indicating the past or present deterioration state of the components, and the production plan information and production performance information acquired by the acquisition unit. An output unit that outputs arrival time information indicating the predicted arrival time, A diagnostic system equipped with the following features.

2. The aforementioned deterioration information indicates the deterioration state of the component at each of several past points in time. The diagnostic system according to claim 1.

3. The aforementioned production equipment is This is a component mounting device that uses a suction nozzle to pick up components and attach them to a substrate. The aforementioned diagnostic system further, The air control mechanism for suctioning air in the adsorption nozzle is equipped with a deterioration identification unit that identifies the deterioration state of the components included in the air control mechanism based on flow rate information relating to the flow rate of air flowing through the air control mechanism, The prediction unit, Information indicating the deterioration state identified by the deterioration identification unit is acquired as deterioration information. The diagnostic system according to claim 1.

4. The aforementioned deterioration identification part is, By estimating the degree of deterioration of the aforementioned component, the deterioration state of the component is identified. The diagnostic system according to claim 3.

5. The aforementioned deterioration identification part is, The system determines whether or not there is an abnormality in the component, and if an abnormality is determined, it estimates the degree of deterioration of the component. The diagnostic system according to claim 4.

6. The aforementioned production performance information is, The system includes mounting count information indicating the number of times a component has been mounted on a substrate by the aforementioned suction nozzle, The prediction unit, Based on the number of times the component is used and the deterioration information, the future deterioration state of the component is estimated. The diagnostic system according to claim 3.

7. The aforementioned specified deterioration state is, The first specified deterioration state requires maintenance of the aforementioned component, or the second specified deterioration state requires preparation for the aforementioned maintenance. A diagnostic system according to any one of claims 1 to 6.

8. The prediction unit, As the future deterioration state of the aforementioned component, the future degree of deterioration is estimated, with a larger value indicating a greater degree of deterioration of the component. Based on the production plan information, the time at which the future degree of deterioration reaches a first threshold corresponding to the first specified deterioration state, or the time at which it reaches a second threshold corresponding to the second specified deterioration state, is predicted. The diagnostic system according to claim 7.

9. (i) Production plan information indicating a production plan for producing mounted substrates, which are substrates with components mounted on them, using production equipment, and (ii) Production performance information indicating the actual production of the mounted substrates using the production equipment, Based on deterioration information indicating the past or present deterioration state of the components included in the production equipment, and the acquired production plan information and production performance information, the timing at which the future deterioration state of the components will reach a predetermined deterioration state is predicted. Output arrival time information indicating the predicted arrival time. Diagnostic methods.

Images