Error factor identification device

Abstract

A fault factor identification device comprises: a memory section configured to store operating acceptance information indicating the acceptance of an operating result of each of the multiple assembly operations performed in a component assembly fixture, as well as operating condition information indicating two or more of the multiple components of the component assembly fixture and assembly order data involved in each of the multiple assembly operations;a specification section configured to specify, if an occurrence rate or occurrence count of a defect in the assembly process exceeds a predetermined value in at least one of the component assembly fixture, component, and assembly order data, the component or assembly order data that is a factor of the defect by comparing two or more of each component's defect participation rate and the assembly order data's defect participation rate based on the operating acceptance information and the operating condition information; and an estimation section configured to estimate the component or assembly order data that is the cause of the defect using a predetermined estimation logic that differs from the comparison of defect participation rates.

Description

Technical field

[0001] The present description relates to a fault factor identification device that identifies a fault factor occurring during an assembly process of a component assembly device. State of the art

[0002] Techniques for the mass production of printed circuit board (PCB) products by processing a PCB on which a circuit pattern is formed are known. A typical example of a PCB processing machine is a component assembly machine that performs a component assembly operation. Generally, the component assembly machine has a component feeder, such as a belt feeder, which feeds components, and a component assembly unit, such as a suction nozzle, which picks up a component from the component feeder and mounts the component onto a PCB. If a defect occurs as a result of a failed assembly operation, the component assembly machine discards the affected component and repeats the assembly operation with a different component.For this reason, the disposal rate of components increases with increasing error rates during assembly, and furthermore, the error rate of printed circuit board products can increase.

[0003] Therefore, in the event of an increase in assembly errors, a countermeasure was implemented whereby the component supply unit or component assembly unit contributing to the errors is serviced and reused, or alternatively, replaced with a spare unit. In this case, by correctly identifying and maintaining the fault-causing unit, it is possible to prevent the occurrence of errors and the disposal of components. Furthermore, it is possible to perform maintenance efficiently and reduce labor costs. Instead of servicing a unit only after an increase in errors has occurred, a preventive measure was also implemented to suppress an increase in errors through regular maintenance. A technical example of identifying a fault factor when assembly errors increase is disclosed in patent literature 1.

[0004] Patent literature 1 discloses a device for estimating the cause of assembly errors, comprising a memory section for the error history configured to summarize and store a status of the occurrence of an error resulting from a failed assembly operation in a component assembly device; a factor setting section configured to determine a first factor and a second factor from the devices and data used for the assembly operation; and a first determination section configured to determine whether an error occurrence status obtained for each individual instance of the second factor under a condition in which an individual instance of the first factor is specified is distorted according to a difference in the second factor.a second determination section configured to determine whether the failure status obtained for each instance of the first factor under a condition specifying a single instance of the second factor is biased according to the difference in the first factor; and a root cause estimation section configured to estimate, based on determination results from the first determination section and the second determination section, a causative instance (factor) that is causing the failure. In this way, the root cause estimation device can estimate the causative element based on a large number of determination results with higher reliability than conventional techniques. Citation list for patent literature

[0005] Patent literature 1: WO 2020 / 188774 Summary of the invention; Technical task

[0006] In the technical example from patent literature 1, however, if assembly errors increase, the source of the errors cannot necessarily be identified. In such a case, there is a risk that no countermeasures will be taken and the situation will remain unaddressed for an extended period. Consequently, the component disposal rate and the printed circuit board failure rate increase, leading to higher production costs and lower production efficiency. Furthermore, since the maintenance target is unknown, maintenance cannot be performed efficiently, and the labor costs required for maintenance increase. The source of assembly errors is not limited to the component feeding unit and the component assembly unit, but can also be another component of the assembly fixture or assembly order data involved in the assembly process.

[0007] One objective of the present description is therefore to provide a fault factor identification device that is able to reduce the cases in which a fault factor is not identified when errors increase during an assembly operation in a component assembly device. Solution to the task

[0008] The present description discloses a fault factor identification device comprising: a memory section configured to store operating acceptance information indicating the acceptance of an operating result of each of the multiple assembly operations performed in a component assembly device, as well as operating state information indicating two or more of a plurality of components of the component assembly device, as well as assembly order data involved in each of the plurality of assembly operations;a specification section configured to specify, when an occurrence rate or occurrence count of a defect in the assembly operation exceeds a predetermined value in at least one of the component assembly fixture, component, and assembly order data, the component or assembly order data that is / are a factor of the defect by comparing two or more of each component's defect participation rate and the assembly order data's defect participation rate, calculated based on the operating acceptance information and the operating condition information; and an estimation section configured to estimate the component or assembly order data that is / are the factor of the defect by using a predetermined estimation logic that differs from the comparison of defect participation rates.

[0009] The present description discloses a technical idea in which "the error factor identification device according to claim 1" is amended in claim 5, as originally filed, to "the error factor identification device according to any one of claims 1 to 4", a technical idea in which "the error factor identification device according to claim 1" is amended in claim 6, as originally filed, to "the error factor identification device according to any one of claims 1 to 4", a technical idea in which "the error factor identification device according to claim 1" is amended in claim 7 in the originally filed version to "the error factor identification device according to any one of claims 1 to 4", a technical idea in which "the error factor identification device according to claim 1" is amended in claim 8 in the originally filed version to "the error factor identification device according to any one of claims 1 to 4",a technical idea in which “the error factor identification device according to one of claims 1 to 3 and 5 to 8” in claim 9 as originally filed is amended to “the error factor identification device according to one of claims 1 to 8”, a technical idea in which “the error factor identification device according to one of claims 1 to 3 and 5 to 8” in claim 10 as originally filed is amended to “the error factor identification device according to one of claims 1 to 9”, a technical idea in which “the error factor identification device according to one of claims 1 to 3 and 5 to 8” in claim 13 as originally filed is amended to “the error factor identification device according to one of claims 1 to 12”, a technical idea,in which “the error factor identification device according to one of claims 1 to 3 and 5 to 8” in the originally filed claim 14 is amended to “the error factor identification device according to one of claims 1 to 12”, and a technical idea in which “the error factor identification device according to one of claims 1 to 3” in the originally filed claim 15 is amended to “the error factor identification device according to one of claims 1 to 14”. Advantageous effects of the invention

[0010] In the disclosed fault factor identification device, the memory section stores the operating acceptance information and the operating state information of each of the assembly operations that have been performed a multitude of times in the component assembly device. Furthermore, the specification section specifies the fault factor by comparing two or more fault participation rates when the occurrence rate or number of occurrences of the fault in the assembly operation exceeds the predetermined value. The estimation section estimates the fault factor using estimation logic that differs from that of the specification section.Since “specify” and “estimate” here correspond to a form of “identify”, the estimation section itself can identify (estimate) the error factor in a case where the specification section cannot identify (specify) the error factor and reduce cases where the error factor is not identified. Brief description of the drawings Fig. Figure 1 is a diagram representing a functional block diagram of a fault factor identification device of an embodiment, and a top view schematically showing a configuration example of a component assembly device in which the fault factor identification device is applied. Fig. Figure 2 is a diagram that displays log data from the component assembly device, including operational acceptance information and operational status information. Fig. Figure 3 is a diagram that illustrates an example of an assembly cycle performed by the component assembly device. Fig. Figure 4 is a diagram of an operating sequence that represents the operation of the fault factor identification device of the embodiment. Fig. Figure 5 is a table that represents the number of operations as a population when, for a component feeder and a component assembly device, each time the component assembly device performs an assembly operation on a predetermined number of printed circuit boards, a failure rate is determined jointly for each individual. Fig. Figure 6 is a table that presents a number of error occurrences, an error occurrence rate, and an error participation rate in a first example. Fig. Figure 7 is a diagram of a sub-operational flow, illustrating details of the specification processing performed by a specification section in step S6 of the operational flow. Fig. 4 is executed. Fig. Figure 8 is an error factor specification diagram that schematically represents the processing contents of the specification processing for the first example. Fig. Figure 9 is a table that presents the number of errors, the error occurrence rate, and the error participation rate for a second example. Fig. Figure 10 is an error factor specification diagram that schematically represents the processing contents of the specification processing for the second example. Fig. 11 is a diagram of a sub-operational flow that illustrates details of the estimation processing performed by an estimation section in step S8 of the operational flow of Fig. 4 is executed and uses a final execution time for maintenance. Fig. Figure 12 is a diagram of a response processing sequence performed by the error factor identification device following the operational sequence of Fig. 4 is executed. Fig. Figure 13 is a diagram of a sub-operational flow of the second estimation processing, which is performed by the estimation section and uses the error occurrence rate. Fig. Figure 14 is a diagram of a sub-operational flow of the third estimation processing, which is performed by the estimation section and uses a maintenance execution interval. Fig. Figure 15 is a diagram of a sub-operational flow of the fourth estimation processing, which is performed by the estimation section and uses the next scheduled maintenance time. Description of embodiments 1. Configuration example for component assembly device 1

[0011] First, a top view is shown in Fig. Figure 1 describes a configuration example for the component assembly device 1, in which the fault factor identification device 8 of an embodiment is used. The component assembly device 1 performs an assembly operation to mount the component onto the printed circuit board K. The horizontal direction from left to right on the drawing surface in Fig. 1 is the X-axis direction in which the printed circuit board K is transported, the horizontal direction from the underside (front) to the top (back) on the drawing surface is the Y-axis direction, and the vertical direction is the Z-axis direction. The component assembly device 1 is configured by assembling a printed circuit board transport device 2, a component feeding device 3, a component transfer device 4, and a control device 5, etc., with a base 10.

[0012] The printed circuit board transport device 2 has a pair of guide rails 21 which serve as a conveyor track for the printed circuit board K. The printed circuit board transport device 2 conveys the printed circuit board K, which is at one loading end (left end in Fig. 1) The printed circuit board K is loaded by a printed circuit board loading device or an external conveying device along the guide rails 21 and conveys the printed circuit board K to a predetermined stop position. The predetermined stop position of the printed circuit board K is essentially defined in the middle of the transport path in the conveying direction. The printed circuit board transport device 2 has a positioning mechanism 22 that pushes the printed circuit board K upwards at the stop position and clamps the printed circuit board K between the positioning mechanism 22 and the guide rails 21. The component transfer device 4 performs an assembly operation on the positioned printed circuit board K. After completion of the assembly operation, the printed circuit board transport device 2 conveys the printed circuit board K from the stop position to an unloading end (right end in Fig. 1) and removes the circuit board K from the assembly device.

[0013] The component feeder 3 is arranged on a front section of the top surface of the base 10 in the Y-axis direction. The component feeder 3 has a plurality of component feed units 31 arranged in the X-axis direction. Each of the component feed units 31 feeds a component during the assembly process and corresponds to a part involved in the assembly process. In this embodiment, a belt feeder is used as the component feed unit 31. The belt feeder guides a carrier belt, in which a plurality of components are arranged in a row, to a feed position on a front surface.The component feeder 31 can be a tray feeder, which uses a tray in which the components are each accommodated in a plurality of receiving sections arranged in a grid pattern, or a rod feeder, in which the components are accommodated in a row within a tubular rod.

[0014] The component transfer device 4 comprises a Y-axis motion body 41, an X-axis motion body 42, a mounting head 43, a rotary tool 44, a plurality of component mounting units 45, a printed circuit board camera 46, a parts camera 47, and the like. The Y-axis motion body 41 consists of an element elongated in the X-axis direction and is driven by a Y-axis drive mechanism (not shown) to move in the Y-axis direction. The X-axis motion body 42 is mounted on the Y-axis motion body 41 and is driven by an X-axis drive mechanism (not shown) to move in the X-axis direction. The mounting head 43 is attached to a front face of the X-axis motion body 42.The assembly head 43, together with the X-axis motion body 42, is driven in two horizontal directions by the drive device and moves to positions above the component feed device 3 and above the circuit board K.

[0015] The rotary tool 44 is rotatably mounted below the assembly head 43. The rotary tool 44 is driven by an R-axis drive mechanism (not shown) to rotate about a vertical central axis. The rotary tool 44 holds a plurality of component mounting units 45 on its underside. Each component mounting unit 45 receives the component at a feed position of the component feeder 31 and mounts the component at a mounting position on the printed circuit board K, thus representing a component involved in the assembly process. In this embodiment, a suction nozzle is used as the component mounting unit 45. The assembly head 43 and the rotary tool 44 hold a plurality of suction nozzles and move between the component feeder 31 and the printed circuit board K, representing a component involved in the assembly process.

[0016] In the Fig. In the example shown in Figure 1, the rotary tool 44 has 12 suction nozzles (the component assembly units 45) arranged at equal intervals from the vertical center axis. The suction nozzle is raised and lowered by a lifting and lowering drive mechanism (not shown) and rotated about a vertical axis by a Q-axis drive mechanism (not shown). The suction nozzle is also selectively supplied with either negative or positive pressure air from an air supply mechanism. Accordingly, the suction nozzle performs an assembly operation in which the component is picked up from the component feed unit 31 and mounted on the printed circuit board K. A plurality of assembly heads 43, a plurality of rotary tools 44, and a plurality of component assembly units 45 are provided, which are exchanged automatically or manually as required.The assembly head 43 can be configured without a rotary tool 44 and with a variety of suction nozzles 45 arranged in a row or in a grid. The component assembly unit 45 is not limited to the suction nozzle and can be a gripper that grasps and picks up the component.

[0017] The printed circuit board camera 46 is arranged side-by-side on the X-axis motion body 42, with the assembly head 43 pointing downwards. The printed circuit board camera 46 images a reference mark attached to the printed circuit board K from above. The captured image data undergoes image processing so that the stop position of the printed circuit board K is precisely determined. The component camera 47 is arranged on the base 10 between the printed circuit board transport device 2 and the component feeder 3, pointing upwards. The component camera 47 captures the component picked up by the component assembly unit 45 from below, while the assembly head 43 moves from the component feeder 3 to the printed circuit board K.The acquired image data undergoes image processing to determine whether the component type is correct or incorrect. Furthermore, the component's position and orientation relative to the component assembly unit 45 are recorded and taken into account during the assembly process. The part camera 47 corresponds to a component involved in the assembly process. Examples of the printed circuit board camera 46 and the part camera 47 are digital imaging devices that incorporate an imaging element, such as a charge-coupled device (CCD) or a complementary metal-oxide semiconductor (CMOS).

[0018] The component transfer device 4 performs the assembly process by repeating an assembly cycle. To describe the assembly cycle in detail, the component transfer device 4 first moves the assembly head 43 to the component feeder 31 and picks up the components using a plurality of component assembly units 45. Next, the component transfer device 4 moves the assembly head 43 to a position above the part camera 47. The part camera 47 then captures images of the component holding states of a plurality of component assembly units 45. Next, the component transfer device 4 moves the assembly head 43 to the printed circuit board K and assembles a plurality of components there. Finally, the component transfer device 4 moves the assembly head 43 back to the component feeder 31, thus completing one assembly cycle.

[0019] The control device 5 is mounted on base 10, and its position is not restricted. The control device 5 is configured as a computer device that includes a CPU and operates with software. The control device 5 can be configured so that multiple CPUs are distributed and interconnected within the device. The control device 5 has a storage device 51 that stores various types of information. The control device 5 is interconnected with the line management device 6. The control device 5 receives assembly order data 52 transmitted by the line management device 6 and stores the assembly order data 52 in the storage device 51. The assembly order data 52 is data used for the assembly process and is generated for each type of printed circuit board K (PCB product).

[0020] The assembly order data 52 includes printed circuit board data relating to the shape and other characteristics of the printed circuit board K and component data relating to the shape and other characteristics of a component to be mounted on the printed circuit board K. There is a multitude of component data, which varies depending on the type and size of the component. Each of the multitude of component data included in the assembly order data 52 can be a factor contributing to an error in the assembly process. This is similar to a case where each of the multitude of component feeder units 31 that make up the component feeder device 3 can be a factor contributing to an error, and a case where each of the multitude of component assembly units 45 that are held by the assembly head 43 can be a factor contributing to an error.

[0021] Furthermore, the assembly order data 52 includes coordinate data of the feed position of the component feed unit 31 and coordinate data of the mounting position on the printed circuit board K on which the component is mounted. The assembly order data 52 also includes data regarding the type and arrangement position of the component feed unit 31 and the component assembly unit 45 to be used, detailed process data of the assembly operation, and the like. The control device 5 controls the printed circuit board transport device 2, the component feed unit 3, and the component transfer device 4 based on the assembly order data 52, thereby repeating the assembly cycle and advancing the assembly process. 2. Line management device 6 and log data LD

[0022] Next, the line management device 6, which manages the component assembly device 1, and the log data LD, which records an operating history of the component assembly device 1, are described. The line management device 6 manages the printed circuit board production line 7, in which a variety of board processing machines, including the component assembly device 1, are arranged. The line management device 6 is configured using a computer device. The line management device 6 has an input device 61, such as a keyboard or touch panel, which receives an instruction, a selection operation, or the like from a worker. Furthermore, the line management device 6 has a display device 62, such as a liquid crystal display, which shows various types of information to the worker.

[0023] The line management device 6 is communicatively connected to the feeder maintenance device 77 and the nozzle maintenance device 78. The feeder maintenance device 77 receives the component feeder unit 31 (belt feeder) and performs maintenance on it. The line management device 6 can record the execution history of the maintenance performed by the feeder maintenance device 77 on the component feeder unit 31. Furthermore, if regular maintenance of the component feeder unit 31 is performed, the line management device 6 can receive a maintenance execution interval and a future execution plan from the feeder maintenance device 77.

[0024] Meanwhile, the nozzle maintenance device 78 receives the component assembly unit 45 (suction nozzle) and performs maintenance work on it. The line management device 6 can record the execution history of the maintenance work performed by the nozzle maintenance device 78 on the component assembly unit 45. If regular maintenance work is performed on the component assembly unit 45, the line management device 6 can also record a maintenance execution interval and a future execution plan from the nozzle maintenance device 78. Maintenance of the component feed unit 31 (belt feeder) and the component assembly unit 45 (suction nozzle) can be performed by the operator. In this case, the operator can enter the execution history of the maintenance into the line management device 6 via the input device 61 and also enter the execution interval and the execution plan for the maintenance.

[0025] The numerous printed circuit board (PCB) processing machines that comprise PCB production line 7—namely, the solder paste printing machine 71, the print inspection machine 72, the component assembly device 1, the visual PCB inspection device 73, and the reflow machine 74—are arranged sequentially from the upstream side. These PCB processing machines are communicatively connected to the line management device 6. The line configuration of PCB production line 7 can be modified in various ways. Furthermore, the line management device 6 can manage multiple PCB production lines 7.

[0026] The line management device 6 receives log data LD, essentially in real time, which records a detailed history of the operating status of each of the board processing machines. The data format of the log data LD can be different for each type of board processing machine or it can be standardized. The log data LD of the component assembly device 1 contains history information for each of the multiple assembly operations performed. The log data LD of the component assembly device 1 can also contain transport history information regarding the transport of the printed circuit board K and exchange history information regarding the exchange of a component.

[0027] Fig. Figure 2 represents an example of protocol data LD received by the line management device 6 from the control device 5 of the component assembly device 1. Each line in Fig. 2 corresponds to a data record LD. The log data LD, which relates to the assembly process of component assembly device 1, has a data format in which six pieces of information are linked together. That is, the log data LD links time information, component type information, ID information of component feed unit 31, ID information of component assembly unit 45, assembly position information, and operational acceptance information. Line numbers 1 to 8 were added for simplicity to distinguish and describe the eight log data LD. Furthermore, the data format of the log data LD of component assembly device 1 can be modified accordingly.

[0028] The time information specifies the point in time (hour:minute:second) at which the assembly operation is performed and is represented, for example, by the point in time at which the lifting of component assembly unit 45, which mounted the component onto the printed circuit board K, ends. The component type information specifies the type of component that is the subject of the assembly operation. The ID information of component feeder unit 31 is information for identifying the individual component feeder unit 31 that fed the component. The ID information of component assembly unit 45 is information for identifying the individual component assembly unit 45 that picked up the component and mounted it onto the printed circuit board K. The ID information of component feeder unit 31 and component assembly unit 45 corresponds to the operating state information, which specifies a component involved in each of the multiple assembly operations.The mounting position information specifies the mounting position on the printed circuit board K where the component is mounted in an XY coordinate system. The operational acceptance information indicates the acceptance (normal or faulty) of a result from the component's mounting process.

[0029] The log data LD in line 1 in Fig. The data relating to the assembly process at time 10:31:02 are as follows: The log data LD indicates that a component of type P1 was supplied by component feeder unit 31 with ID information F1 and mounted by component assembly unit 45 with ID information N1 at the mounting position (x1, y1) on the printed circuit board K. Furthermore, the log data LD indicates that the result of the component assembly process is normal.

[0030] Furthermore, four log data entries (LD) in lines 1 to 4 show that four components of type P1 were supplied by component feeder 31 with ID information F1 and mounted at four locations on the printed circuit board (PCB) K by four component assembly units 45 with ID information N1 to N4. The log data entries (LD) in lines 1, 2, and 4 indicate that the results of the assembly operation at mounting positions (x1,y1), (x2,y2), and (x4,y4) on PCB K are normal. However, the log data entry (LD) in line 3 indicates that the result of the assembly operation at mounting position (x3,y3) on PCB K is faulty. For mounting position (x3,y3), the assembly cycle for recovery is set, and the assembly operation for the component of type P1 is repeated.

[0031] Two log entries (LD) in lines 5 and 6 indicate that two P2 components were fed by component feeder 31 with ID information F2 and mounted at two locations on the printed circuit board (PCB) K by two component assembly units 45 with ID information N1 and N2. It is further indicated that the assembly results at mounting positions (x5,y5) and (x6,y6) on PCB K are normal. Additionally, two log entries (LD) in lines 7 and 8 indicate that two P3 components were supplied by component feeder 31 with ID information F3 and mounted at two locations on PCB K by two component assembly units 45 with ID information N3 and N4. It is further indicated that the assembly results at mounting positions (x7,y7) and (x8,y8) on PCB K are normal.

[0032] In component assembly device 1, an error occasionally occurs that is attributable to a failed assembly operation. Examples of assembly operation errors include the following cases 1) to 7). 1) A case in which no image data is captured by the parts camera 47. 2) A case in which the image data is not subjected to suitable image processing. 3) A case in which image processing reveals that the component assembly unit 45 is not holding a component. 4) A case in which image processing reveals that the component type is incorrect. 5) A case in which image processing reveals a significant error in the component's positioning, making assembly impossible. 6) A case in which the component falls out of the component assembly unit 45 while the assembly head 43 moves from the part camera 47 to the circuit board K. 7) A case in which the component is not mounted on the circuit board K while it is held by the component assembly unit 45.

[0033] In addition to the cases mentioned above, there is a defect that is detected based on a test result from the visual printed circuit board inspection device 73. Specifically, the visual printed circuit board inspection device 73 detects a defect if there is an excessive error in the actual mounting position of the component mounted on the printed circuit board K, if there is an excessive error in the orientation (rotation in a horizontal plane) of the component, or if the component is mounted at an excessive angle. Subsequently, the visual printed circuit board inspection device 73 transmits the log data LD, which contains the inspection result of the defect, including the mounting position information of the component identified as defective, to the line management device 6.The line management device 6 extracts the log data LD of the corresponding component assembly device 1 based on the assembly position information of the component that was identified as faulty and is contained in the log data LD of the visual printed circuit board inspection device 73, and rewrites the operating acceptance information from "normal" to "fault".

[0034] The error factors in the cases described above vary, including a case where the factor is the component feed unit 31 or the component assembly unit 45 (part of the component assembly fixture 1), and a case where the factor is the assembly order data 52. Furthermore, components such as the assembly head 43, the rotary tool 44, and the part camera 47 can be the error factor. To improve efficiency and reduce the workload involved in handling an increase in such errors, an error factor identification device 8 is used. 3. Configuration and function of the fault factor identification device 8

[0035] Next, the configuration and function of the fault factor identification device 8 of the embodiment will be described with reference to a functional block diagram in Fig. The fault factor identification device 8 is described in Section 1. It is configured as a management function section of the line management device 6. However, the fault factor identification device 8 is not limited to this configuration and can be configured within the control device 5 of the component assembly device 1 or using another computer device. The fault factor identification device 8 comprises four functional sections: the memory section 81, the specification section 82, the estimation section 83, and the notification section 84.

[0036] Memory section 81 stores log data LD of component assembly device 1 in a memory (not shown) or the like. In other words, memory section 81 stores the operating acceptance information and the operating condition information. The operating acceptance information is contained in each of the Fig. The operating condition information is information that specifies two or more of the multiple components of the component assembly device 1 and the assembly order data 52 involved in each of the multiple assembly operations. Information that specifies the component feed unit 31 and the component assembly unit 45 in the operating condition information is included in each of the Fig. The protocol data LD to be displayed contains information about the assembly head 43, the rotary tool 44, and the parts camera 47 in the operating status information. This information is captured from protocol data LD, which contains the exchange history information of the components of the component assembly device 1. The assembly order data 52 in the operating status information is already stored in the storage device 51.

[0037] Specification section 82 performs a specification processing in which the error factor is specified by using as a trigger condition the fact that the error occurrence rate EC of the assembly operation exceeds a predetermined value E1 in at least one of the following component assembly fixtures, one of the components of component assembly fixture 1, and assembly order data 52. Additionally, one or more combinations of the total number of operations of component assembly fixture 1, the number of operations for each individual component feeder 31, the number of operations for each individual component assembly unit 45, and the number of operations for each element of the component data of the assembly order data 52 can be used as the population in calculating the error occurrence rate EC.Furthermore, a predetermined value E1 can be set, for example, 0.03% for the total number of operations of component assembly device 1 and 0.1% for the number of other operations.

[0038] Specification section 82 can calculate the failure rate EC for each of the multiple components and the assembly order data 52, each time the component assembly device 1 performs the assembly operation on a predetermined number of printed circuit boards K, or each time the operating time of the component assembly device 1 is a predetermined time. In this way, it is possible to align the time periods of the populations for calculating the failure rate EC of each of the multiple components and the assembly order data 52, even if the sizes of the multiple populations differ.

[0039] Furthermore, specification section 82 for the multiple components and assembly order data 52 can individually calculate the failure rate EC for each instance whenever the number of participants in the assembly operation reaches a predetermined number. For example, specification section 82 can individually calculate the failure rate EC each time each of the multiple component feed units 31 operates 20,000 times, and the failure rate EC each time each of the multiple component assembly units 45 operates 10,000 times. In this way, it is possible to align the population sizes for calculating the failure rate EC of each of the multiple components and assembly order data 52 to a predetermined number, even as the time periods of the multiple populations vary.

[0040] Instead of the error occurrence rate (EC), specification section 82 can perform specification processing using the fact that an error occurrence count exceeds a predetermined value as a trigger condition. When determining the error occurrence count, specification section 82 defines a population in the same way as when calculating the error occurrence rate (EC), determines the error occurrence count within the population, and does not determine a cumulative occurrence count across populations.

[0041] Specification section 82 performs specification processing to compare the defect participation rates ER and specifies the defect factor. Specification section 82 first calculates two or more defect participation rates ER of each individual component and defect participation rates ER of the assembly order data 52 based on the operational acceptance information and the operational state information. Next, given that the component or assembly order data 52 with the higher defect participation rate ER is taken as the first factor and an individual instance of the first factor is specified, specification section 82 determines whether the defect participation rate ER is biased according to a difference in an individual instance of another type of component or assembly order data 52, which is taken as the second factor.If there is no distortion, specification section 82 identifies the specified single instance as the first factor as the error factor.

[0042] If bias exists, the error factor is not specified. In this case, specification section 82 further determines whether the error participation rate (ER) is biased according to a difference in the single instance of the first factor under a condition where a single instance of the second factor, in which the error occurred in a biased manner, is specified. If no bias exists, specification section 82 identifies the specified single instance of the second factor as the error factor.During specification processing, the error factor is not specified if the error participation rate ER is skewed according to the difference in the single instance of the second factor under the condition that the single instance of the first factor is specified, and the error participation rate ER is skewed according to the difference in the single instance of the first factor under the condition that the single instance of the second factor is specified.

[0043] The component feeder unit 31 and the component assembly unit 45 can be used as options for the first and second factors, respectively. Furthermore, with regard to the assembly order data 52, several elements of the component data can be considered as distinct individual instances and used as options. Additionally, the assembly head 43 and the rotary tool 44 can be used as options if log data LD, including exchange history information, is available and multiple individuals are used. Moreover, with regard to the part camera 47, if imaging is performed by switching between multiple imaging conditions, the multiple imaging conditions can be considered as distinct individual instances and used as options.Furthermore, the error participation rate ER is represented by a ratio where a total number of error occurrences is used as the denominator and a number of error occurrences that occurred in each of the multiple individual instances included in the first factor or the second factor is used as the numerator.

[0044] In this embodiment, whether the error participation rate ER is distorted is determined as follows. Specifically, specification section 82 stipulates that the error participation rate ER is distorted if the error participation rate ER of any of the multiple individual instances included in the first factor or the second factor is equal to or greater than the predetermined participation rate E2. Furthermore, specification section 82 stipulates that the error participation rate ER is not distorted if the error participation rate ER of any of the multiple individual instances included in the first factor or the second factor is less than the predetermined participation rate E2. For example, the predetermined participation rate E2 could be set to 80%. The specification processing of specification section 82 is described in detail in the subsequent description of operation, using an assumed case.

[0045] Estimation section 83 uses estimation logic that differs from the specification processing of specification section 82; that is, estimation logic that differs from the comparison of the error participation rates ER to estimate the components or assembly order data 52 that represent the error factor. In the embodiment, estimation section 83 operates only if specification section 82 cannot specify the error factor. For example, estimation section 83 operates when specification section 82 cannot specify the error factor in the specification processing where component feeder unit 31 and component assembly unit 45 are the first and second factors, respectively. Estimation section 83 can operate regardless of the success or failure of specification processing in specification section 82.

[0046] In an example of the estimation logic, estimation section 83 compares the last maintenance execution times performed on the component feeder unit 31 and the component assembly unit 45, which are considered potential factors of the failure, and the unit with the older last execution time is considered a factor of the failure. That is, estimation section 83 performs an estimate based on a basic idea (an idea according to an empirical rule) that "a unit that has not been serviced for a long time since maintenance is highly likely to be a factor of a failure due to degradation of performance." Even if the assembly head 43 or the rotary tool 44 is selected as one of the first and second factors, respectively, estimation section 83 can perform the estimate by comparing the last maintenance execution times.The estimation logic of estimation section 83 can be modified as described below.

[0047] Notification section 84 provides notification of the fault factor specified by specification section 82 and the fault factor estimated by estimation section 83. Notification section 84 can issue the notification using the display device 62 and, as an additional notification method, send a notification to a worker's mobile device via wireless communication. Furthermore, any two of the component feeder unit 31, component assembly unit 45, assembly head 43, and rotary tool 44 can be selected as the first and second factors. In this case, notification section 84 issues a notification to perform maintenance on the fault factor specified by specification section 82 and the fault factor estimated by estimation section 83.

[0048] It cannot be said that the functions of specification section 82 and estimation section 83 are perfect, and it is difficult to rule out the possibility that the failure factor might be faulty. That is, even if maintenance is performed on a first component feeder unit 31 and component assembly unit 45 corresponding to the failure factor reported by notification section 84, and the unit is reused, there are cases in which the failure rate EC is not improved to a value equal to or less than the predetermined value E1. In a case where the failure rate EC is not improved, estimation section 83 corrects the estimate so that a second component feeder unit 31 and component assembly unit 45, which is in continuous use, is considered the failure factor.Then, notification section 84 issues a notification to perform maintenance on the corrected error factor. The function of estimation section 83 to correct the error factor can be omitted. 4. Operation of the fault factor identification device 8

[0049] Next, the operation of the fault factor identification device 8 of the embodiment will be described with reference to the Fig. 3 to 12 are described using a hypothetical case. Before describing the operation, the assembly process on the printed circuit board K is assumed as follows. That is, it is assumed that a total of 11 components are mounted on the printed circuit board K, including four components of type P1, two components of type P2, two components of type P3, two components of type P4, and one component of type P5. To fulfill this assumption, the component assembly device 1 performs the operation described in Fig. The assembly process shown in section 3 is carried out with three assembly cycles.

[0050] As in Fig. As shown in Figure 3, the component assembly device 1 uses a first nozzle N1, a second nozzle N2, a third nozzle N3, and a fourth nozzle N4, corresponding to four component assembly units 45 (suction nozzles). Furthermore, the component assembly device 1 uses five component feed units 31 (belt feeders), shown in brackets. Specifically, the component assembly device 1 uses a first feeder F1 that feeds a component of type P1, a second feeder F2 that feeds a component of type P2, a third feeder F3 that feeds a component of type P3, a fourth feeder F4 that feeds a component of type P4, and a fifth feeder F5 that feeds a component of type P5.

[0051] In a first assembly cycle, the component assembly device 1 performs an assembly operation in which four components of type P1 are picked up from the first feeder device F1 using the first nozzle N1 to the fourth nozzle N4 and mounted onto the printed circuit board K. Furthermore, in a second assembly cycle, the component assembly device 1 performs an assembly operation in which two components of type P2 are picked up from the second feeder device F2 using the first nozzle N1 and the second nozzle N2, and two components of type P3 are picked up from the third feeder device F3 using the third nozzle N3 and the fourth nozzle N4.Furthermore, in a third assembly cycle, the component assembly device 1 performs an assembly operation in which two components of type P4 are picked up from the fourth feeder device F4 using the first nozzle N1 and the second nozzle N2, and one component of type P5 is picked up from the fifth feeder device F5 using the third nozzle N3. The component assembly device 1 does not use the fourth nozzle N4 in the third assembly cycle.

[0052] Furthermore, it is assumed that specification section 82 selects the component feeder unit 31 and the component assembly unit 45 as the first and second factors, respectively. It is also assumed that specification section 82 calculates the failure rate EC for each individual component feeder unit 31 and each individual component assembly unit 45 each time the component assembly device 1 performs the assembly operation on 5000 printed circuit boards K, and that the predetermined value E1, as the trigger condition for specification section 82 to execute the specification processing, is set to 0.1%. Additionally, it is assumed that the predetermined participation rate E2, used to determine whether the failure participation rate ER is biased, is set to 80%.

[0053] Based on the above assumption, the functioning of the error factor identification device 8 is described with reference to a [reference to a document]. Fig. The functional sequence shown in section 4 is described. In step S1 of Fig. 4. Component assembly device 1 executes the assembly operations of the first through third assembly operations sequentially, then switches over the printed circuit board K and continues the assembly operation. In the next step S2, memory section 81 captures and stores the log data LD relating to each of the assembly operations of component assembly device 1. As described above, the log data LD contains the operating acceptance information and the operating condition information for each assembly operation. The execution frequency of the memory processing can be at each time an assembly operation is completed, at each time an assembly cycle (three to four assembly operations) is completed, or at each time 11 assembly operations are completed on a printed circuit board K.

[0054] In the next step, S3, specification section 82 determines whether the calculation time for the failure rate EC has been reached and branches the workflow accordingly. The calculation time has not yet been reached if the assembly operations on 5000 printed circuit boards K are not yet completed after the start of production of the printed circuit board K or after the previous calculation time. In this case, specification section 82 returns to the workflow in step S1. The work loop of steps S1 to S3 is then repeated until the calculation time is reached, and log data LD is collected. When the calculation time in step S3 is reached, specification section 82 continues the workflow to step S4.

[0055] In step S4, specification section 82 jointly calculates the fault occurrence rate EC for each individual component feed unit 31 and each individual component assembly unit 45. Fig. Figure 5 illustrates the number of operations as a population for calculating the failure rate EC. As shown in the figure, the total number of assembly operations for component assembly unit 1 is 55,000 (= 5,000 printed circuit boards × 11 operations). The number of operations for each individual component feeder unit 31 is 20,000 for the first feeder, 10,000 for the second feeder F2, the third feeder F3, and the fourth feeder F4, and 5,000 for the fifth feeder F5. The number of operations for each individual component assembly unit 45 is 15,000 for the first nozzle N1, the second nozzle N2, and the third nozzle N3, and 10,000 for the fourth nozzle N4.

[0056] Here, in a first example, which is in Fig. Figure 6 shows that a total of 25 defects occur in the component assembly device 1. The details of the defect occurrence count for each individual component feeder 31 are such that the first feeder has one occurrence, the second feeder F2 has 22 occurrences, the third feeder F3 has one occurrence, the fourth feeder F4 has no occurrences, and the fifth feeder F5 has one occurrence. Specification section 82 calculates the defect occurrence rate EC by dividing this defect occurrence count by the number of in Fig. The 5 processes represented were divided by the population. The specific error rates EC are 0.01% for the first feeder, 0.22% for the second feeder F2, 0.01% for the third feeder F3, 0% for the fourth feeder F4, and 0.02% for the fifth feeder F5 (values ​​below 0.01% are rounded).

[0057] The details of the number of failures for each individual component assembly unit 45 are such that the first nozzle N1 has nine occurrences, the second nozzle N2 has 13 occurrences, the third nozzle N3 has two occurrences, and the fourth nozzle N4 has one occurrence. Specification section 82 calculates the failure rate EC by dividing this number of failures by the number of in Fig. The processes represented in Figure 5 are divided by the population. The specific error rates (EC) are 0.06% for the first nozzle N1, 0.09% for the second nozzle N2, 0.01% for the third nozzle N3, and 0.01% for the fourth nozzle N4.

[0058] In the next step, S5, specification section 82 determines whether the error rate EC exceeds a predefined value E1 (= 0.1%) and branches the operating sequence. If the number of errors is low and no error rate EC exceeds a predefined value E1, specification section 82 returns to step S1. Steps S1 through S5 are then repeated. In the first example, the error rate EC of the second feeding device F2 is 0.22% and exceeds the predefined value E1, and specification section 82 forwards the operating sequence to step S6. Furthermore, it is assumed that the predefined value E1 for the total number of operations of component assembly device 1 can be set to 0.03% as a trigger condition for specification section 82 to execute the specification processing.In this case, the failure rate EC of component assembly device 1 is 0.05% (= 25 / 55,000 × 100) and exceeds the specified value E1, so that specification section 82 forwards the workflow to step S6.

[0059] In step S6, specification section 82 performs specification processing to specify the fault factor. If the second feeder F2 with the highest number of fault occurrences is simply identified as the fault factor, there is a possibility that it is faulty. Therefore, specification section 82 performs the processing in a sub-operation in Fig. The specification processing shown in section 7 is shown. In step S11 in Fig. 7. Specification section 82 calculates the error participation rate ER for each individual component feed unit 31 and each individual component assembly unit 45. As in Fig. As shown in Figure 6, in the first example, the error participation rate (ER) for each individual component feeder is 31%: 4% for the first feeder, 88% (= 22 / 25 × 100) for the second feeder F2, 4% for the third feeder F3, 0% for the fourth feeder F4, and 4% for the fifth feeder F5. Furthermore, the error participation rate (ER) for each individual component assembly unit is 45%: 36% for the first nozzle N1, 52% (= 13 / 25 × 100) for the second nozzle N2, 8% for the third nozzle N3, and 4% for the fourth nozzle N4.

[0060] In the next step S12, specification section 82 designates a single component as the first factor, which has a high defect participation rate (ER). In the first example, specification section 82 selects the component feeder 31 as the first factor, selects the component assembly unit 45 as the second factor, and designates the second feeder F2, with a high defect participation rate (ER) of 88%, as the first factor. In the next step S13, specification section 82 determines the presence or absence of distortion based on a difference in the single instance of the second factor, given that the second feeder F2 is specified. Given that the second feeder F2 is specified, the defect occurrence count is 22, with the first nozzle N1 having nine occurrences and the second nozzle N2 having 13 occurrences.

[0061] Specification section 82 calculates the error participation rate ER for each individual case of the second factor, provided that the second feeding device F2 is specified. As shown in brackets in Fig. As shown in Figure 6, the failure participation rate ER of the first nozzle N1, as the second factor, is 38% (= 9 / 22 × 100) and the failure participation rate ER of the second nozzle N2 is 62% (= 13 / 22 × 100). Both failure participation rates ER are below the predetermined participation rate E2 (= 80%), and there is no bias due to the difference between individuals as the second factor. Therefore, the sub-operational sequence branches to step S14.

[0062] In step S14, specification section 82 designates the specified unit as the first factor, i.e., the second feeding device F2 as the fault factor. It is, of course, highly likely that the second feeding device F2 with the highest number of fault occurrences is the fault factor. However, it cannot be ruled out that one of the component assembly units 45 used in combination is the fault factor. Here, it is noted that the faults are not limited to either the first nozzles N1 or the second nozzles N2, but occur in both. It is extremely rare for the first nozzle N1 and the second nozzle N2 to experience a simultaneous degradation in performance and cause a fault.Therefore, specification section 82 can state that the occurrence of the fault is due to a deterioration in the performance of the second feeding device F2, and there is almost no possibility that the specification result is faulty.

[0063] The processing content of the specific processing is schematically represented in an error factor specification diagram in Fig. Figure 8 illustrates this. In the defect factor specification diagram, the vertical axis represents the defect participation rate ER of the component feeder 31, and the horizontal axis represents the defect participation rate ER of the component assembly unit 45. The defect factor specification diagram is divided into four regions by a predetermined participation rate E2 (= 80%), indicated by a dashed line. Specifically, the first region, A1, is defined as the region in which the defect participation rate ER of both the component feeder 31 and the component assembly unit 45 is less than the predetermined participation rate E2. The second region, A2, is defined as the region in which the defect participation rate ER of the component feeder 31 is equal to or greater than the predetermined participation rate E2, and the defect participation rate ER of the component assembly unit 45 is less than the predetermined participation rate E2.A region in which the defect participation rate ER of component feeder 31 is less than the predetermined participation rate E2 and the defect participation rate ER of component assembly unit 45 is equal to or greater than the predetermined participation rate E2 is the third region A3. A region in which the defect participation rates ER of component feeder 31 and component assembly unit 45 are both equal to or greater than the predetermined participation rate E2 is the fourth region A4.

[0064] Specification section 82 calculates the defect participation rate ER of the first nozzle N1 and the defect participation rate ER of the second nozzle N2, provided that the second feeder F2 is specified as described above. The calculation results are represented as P1(F2,N1) and P2(F2,N2) in the defect factor specification diagram. For example, P1(F2,N1) is represented at the intersection of 88% of the defect participation rate ER of the second feeder F2 and 38% of the defect participation rate ER of the first nozzle N1, provided that the second feeder F2 is specified. As shown in the drawing, specification section 82 can specify one of the component feeder units 31 as a defect factor if diagrams such as P1(F2,N1) and P2(F2,N2), which indicate the processing content of specification section 82, are included in the second area A2.Taking into account the duality between the component feed unit 31 and the component assembly unit 45, the specification section 82, when the plots showing the processing content of the specification section 82 are included in the third area A3, can specify one of the component assembly units 45 as the factor of the error.

[0065] As an application example, a case is now described in which the first and second factors are toggled and selected. In step S12, specification section 82 can select component assembly unit 45 as the first factor, component feed unit 31 as the second factor, and specify the second nozzle N2, which has a high defect participation rate ER of 52%, as the first factor. Then, in the next step S13, specification section 82 determines the presence or absence of distortion corresponding to a difference in the individual as the second factor, given that the second nozzle N2 is specified. Given that the second nozzle N2 is specified, the defect occurrence count is 13.The number of failures for the second feeder F2, as the second factor, is 13 with a failure participation rate ER of 100%, and the number of failures for the first feeder F1 and the fourth feeder F4 is 0 with a failure participation rate ER of 0%. This means that the failure participation rate ER of the second feeder F2 is equal to or greater than the predetermined participation rate E2 and is therefore distorted. Consequently, the sub-operation branches to step S15. The processing contents are represented as Q1(N2,F1), Q2(N2,F2), and Q3(N2,F4) in the failure factor specification diagram. Fig. 8 shown.

[0066] In step S15, specification section 82 defines the second feeder F2, which is an individual, as the second factor in which the fault occurred in a distorted manner. In the next step S16, specification section 82 determines the presence or absence of distortion according to a difference in the individual as the first factor, given that the second feeder F2 is specified. Given that the second feeder F2 is specified, the fault occurrence count is 22. The fault participation rate ER of the first nozzle N1, as the first factor, is 38%, and the fault participation rate ER of the second nozzle N2 is 62%. Therefore, the fault participation rate ER is not distorted, and the sub-operational sequence branches to step S17.

[0067] In step S17, specification section 82 specifies the specified individual as the second factor, i.e., the second feeder F2 as the fault factor. As described above, specification section 82 can obtain the same specification result as in the first example, even if the specification processing is executed where the first and second factors are reversed in the application example. Executing step S14 or step S17 terminates the sub-operational flow of the specification processing.

[0068] Next, the specification processing of specification section 82 is described when a Fig. The second example shown in Figure 9 is used as the target. In the second example, it is assumed that a total of 25 defects occur in the component assembly device 1. The details of the defect occurrence count for each individual component feeder 31 are such that the first feeder has one occurrence, the second feeder F2 has 22 occurrences, the third feeder F3 has one occurrence, the fourth feeder F4 has no occurrences, and the fifth feeder F5 has one occurrence. Therefore, the defect occurrence rate EC for each individual component feeder 31, calculated by specification section 82 in step S4, is 0.01% for the first feeder, 0.22% for the second feeder F2, 0.01% for the third feeder F3, 0% for the fourth feeder F4, and 0.02% for the fifth feeder F5.

[0069] Furthermore, the details of the number of failures for each individual component assembly unit 45 are such that the first nozzle N1 has two failures, the second nozzle N2 has 20 failures, the third nozzle N3 has two failures, and the fourth nozzle N4 has one failure. Therefore, the failure rate EC for each individual component assembly unit 45, calculated by specification section 82, is 0.01% for the first nozzle N1, 0.13% for the second nozzle N2, 0.01% for the third nozzle N3, and 0.01% for the fourth nozzle N4.

[0070] In the next step S5, the error rate EC of the second feeding device F2 is 0.22%, which is equal to or greater than the specified value E1, so that specification section 82 forwards the operating sequence to step S6. In step S11 of Fig. 7, which corresponds to the specification processing of step S6, the error participation rate ER for each individual component feeder 31, calculated by specification section 82, is 4% for the first feeder, 88% (= 22 / 25 × 100) for the second feeder F2, 4% for the third feeder F3, 0% for the fourth feeder F4, and 4% for the fifth feeder F5. Furthermore, the error participation rate ER for each individual component assembly unit 45 is 8% for the first nozzle N1, 80% (= 20 / 25 × 100) for the second nozzle N2, 8% for the third nozzle N3, and 4% for the fourth nozzle N4.

[0071] In the next step S12, specification section 82 selects the component feeder 31 as the first factor, the component assembly unit 45 as the second factor, and the second feeder F2, which exhibits a high defect rate, as the first factor. In the next step S13, specification section 82 calculates the defect rate ER of the first nozzle N1 as the second factor at 9% (= 2 / 22 × 100) and the defect rate ER of the second nozzle N2 at 91% (= 20 / 22 × 100), provided that the second feeder F2 is specified. Since the defect rate ER of the second nozzle N2 is preset to be equal to or greater than the predetermined defect rate E2 (= 80%), specification section 82 proceeds to step S15.

[0072] In step S15, specification section 82 specifies the second nozzle N2, which is a single instance of the second factor where the fault occurred in a distorted manner. In the next step S16, specification section 82 calculates the fault participation rate ER of the second feeder F2 as the first factor at 100% (= 22 / 22 × 100) and the fault participation rates ER of the first feeder F1 and the fourth feeder F4 at 0%, given that the second nozzle N2 is specified. Therefore, the fault participation rate ER of the second feeder F2 is distorted to be equal to or greater than the predetermined participation rate E2, and specification section 82 proceeds to step S18. In step S18, specification section 82 determines that the fault factor cannot be specified, and the sub-operation ends.

[0073] The processing content of the specific processing in the second example is schematically represented in an error factor specification diagram in Fig. Figure 10 illustrates this. In particular, the processing contents of steps S12 and S13 are represented by specification section 82 as P4(F2,N1) and P5(F2,N2) in the error factor specification diagram. Furthermore, the processing contents of step S16 are represented by specification section 82 as Q4(N2,F1), Q5(N2,F2), and Q6(N2,F4) in the error factor specification diagram. As shown in the drawing, specification section 82 cannot specify the error factor if plots such as P5(F2,N2) and Q5(N2,F2), which represent the processing content of specification section 82, are contained in the fourth area, A4.

[0074] Furthermore, there are cases in which a relatively large number of defects are distributed across multiple component feed units 31 and multiple component assembly units 45. In this case, plots showing the processing content of specification section 82 are included in the first area A1, and specification section 82 cannot specify the defect factor. Such a case occurs, for example, when multiple units among multiple component feed units 31 and multiple component assembly units 45 simultaneously experience performance degradation and cause a defect. In the case shown, specification section 82 cannot specify the defect factor, but estimation section 83 can estimate a first unit that is a defect factor.After the first unit has been serviced and reused, or a replacement unit has been used in place of the first unit, specification section 82 can specify a second unit in the next specification processing, which is a factor of the fault.

[0075] Back to step S7 in Fig. 4. Estimation section 83 branches the operational sequence depending on whether specification section 82 has specified the error factor. If specification section 82 was able to specify the error factor, estimation section 83 skips step S8 and continues with the operational sequence in step S9. If, however, specification section 82 was unable to specify the error factor, estimation section 83 executes the sub-operation in step S8. Fig. 11 shown estimation processing.

[0076] In step S21 of Fig. In step 11, estimation section 83 extracts the component feeder unit 31 and the component assembly unit 45 that are likely to be the source of the failure. In the second example, estimation section 83 extracts, for instance, the second feeder unit F2 and the second nozzle N2, where the failure occurred in a distorted manner. In the next step S22, estimation section 83 records the last maintenance execution times on both units. That is, estimation section 83 records the last execution time MAF of maintenance on the second feeder unit F2 from the feeder unit maintenance device 77. Additionally, estimation section 83 records the last execution time MAN of maintenance on the second nozzle N2 from the nozzle maintenance device 78.

[0077] In the next step S23, the estimation section 83 switches the sub-process to step S24 if the last execution time MAF of the second feeder F2 is older than the last execution time MAN of the second nozzle N2; otherwise, it switches the sub-process to step S25. In step S24, the estimation section 83 estimates that the second feeder F2 (component feeder 31) is the fault factor. Furthermore, in step S25, the estimation section 83 estimates that the second nozzle N2 (component assembly unit 45) is the fault factor. That is, the estimation section estimates the unit with the older last maintenance execution time as the fault factor. This concludes the sub-process of the estimation.

[0078] Returning to step S9 in Fig. 4. Notification section 84 issues a notification to perform maintenance on the fault factor specified by specification section 82. If specification section 82 could not specify the fault factor, notification section 84 issues a notification to perform maintenance on the fault factor estimated by estimation section 83. Then the process ends in Fig. 4 shown operating sequence, and the fault factor identification device 8 moves with the in Fig. 12 response processing to be presented. In the response processing, an example case is described in which notification section 84 issues a notification to carry out maintenance work on the second feeder F2 (first), which is indicated by the estimation result of estimation section 83 from the second feeder F2 (first) and the second nozzle N2 (second), where the fault occurred in a distorted manner.

[0079] In step S31 of the Fig. In the response processing shown in step 12, the worker interrupts the operation of component assembly device 1 and removes the second feeder F2, for which maintenance has been reported. In the next step S32, the worker carries the second feeder F2 to the feeder maintenance device 77 and performs the maintenance. In the next step S33, the worker mounts the second feeder F2, for which the maintenance has been completed, onto component assembly device 1, uses the second feeder F2 again, and resumes operation of component assembly device 1. In the next step S34, steps S1 to S4 are repeated. Fig. 4 is executed, and specification section 82 calculates the failure rate EC for each individual second feeding device F2. Estimation section 83 determines whether the calculated failure rate EC has been improved so that it is equal to or less than the predetermined value E1 (= 0.1%), and branches the operating sequence.

[0080] In many cases, the fault factor is correctly specified, appropriate maintenance is performed, and the fault occurrence rate (EC) improves. Therefore, component assembly device 1 can continue the stable assembly process. This concludes the response processing sequence. On the other hand, in step S35, if the fault occurrence rate (EC) is not improved, estimation section 83 corrects the estimate so that the continuously used second nozzle N2 (the second one) is considered the fault factor. In the next step, S36, notification section 84 issues a notification to perform maintenance on the second nozzle N2, which is the corrected fault factor. The response processing sequence then ends.

[0081] The operator removes the second nozzle N2, for which maintenance has been reported, from component assembly device 1 and initiates the nozzle maintenance device 78 to perform the maintenance. The operator then reinstalls the second nozzle N2, for which maintenance has been completed, onto component assembly device 1, reuses the second nozzle N2, and resumes operation of component assembly device 1. Since maintenance is performed on both the second feed device F2 and the second nozzle N2, which are considered potential sources of failure, the occurrence status of the failure is improved in most cases.

[0082] In an alternative procedure from step S33, the operator can mount a replacement sixth feeder F6 on component assembly fixture 1 in place of the second feeder F2, for which maintenance was reported, and resume operation of component assembly fixture 1. In this case, if the fault factor is the second feeder F2, the fault occurrence rate EC is improved, and the operation continues from step S34 to the end. If the fault factor is not the second feeder F2, the fault occurrence rate EC is not improved, and the operation proceeds from step S34 to step S35. In step S35, estimation section 83 corrects the estimate so that the continuously used second nozzle N2 is considered the fault factor. In the next step S36, notification section 84 issues a notification to perform maintenance on the second nozzle N2.This concludes the response processing process. 5. Modified form of the estimation processing of estimation section 83

[0083] The estimation processing of estimation section 83 can be split into a second estimation processing, which is in Fig. As shown in 13, a third estimation processing method is used in Fig. 14 is shown, and a fourth estimation processing, which is in Fig. The figure shown in section 15 can be modified. In the second to fourth estimates, steps S21, S24, and S25 are the same as in the figure referenced in section 15. Fig. The estimate described in 11 is changed, and the processing contents of steps S22 and S23 are modified.

[0084] In step S22A of the Fig. In the second estimation processing shown in Figure 13, estimation section 83 captures the failure rate ECF of the second feeder F2 (component feeder 31) and the failure rate ECN of the second nozzle N2 (component assembly unit 45) from specification section 82. In the next step S23A, estimation section 83 continues the sub-processing sequence to step S24 if the failure rate ECF of the second feeder F2 is higher than the failure rate ECN of the second nozzle N2; otherwise, it continues the sub-processing sequence to step S25. That is, estimation section 83 performs an estimation based on the basic idea that "a unit that has a high failure rate has a high probability of being a factor in a failure."Estimation section 83 can record the number of failures from specification section 82 instead of the failure rates (ECF, ECN), compare the number of failures and continue the sub-operational process to step S24 or step S25.

[0085] In step S22B of the Fig. In the third estimation process shown in Figure 14, estimation section 83 retrieves the execution interval MBF for the maintenance of the second feeder F2 (component feeder unit 31) from the feeder maintenance device 77. Furthermore, estimation section 83 retrieves the execution interval MBN for the maintenance of the second nozzle N2 (component assembly unit 45) from the nozzle maintenance device 78. In the next step S23B, estimation section 83 switches the sub-operation sequence to step S24 if the execution interval MBF for the maintenance of the second feeder F2 is shorter than the execution interval MBN for the maintenance of the second nozzle N2; otherwise, it switches the sub-operation sequence to step S25.That is to say, Estimation Section 83 performs an estimate based on the basic idea that "short-interval maintenance is recommended for a unit that is prone to performance degradation and is highly likely to be a factor in failure".

[0086] In step S22C of the Fig.In the fourth estimation process shown in Figure 15, estimation section 83 records the next scheduled time MCF for servicing the second feeder F2 (component feeder 31) from the feeder maintenance device 77. Furthermore, estimation section 83 records the next scheduled time MCN for servicing the second nozzle N2 (component assembly unit 45) from the nozzle maintenance device 78. In the next step S23C, estimation section 83 switches the sub-operation sequence to step S24 if the next scheduled time MCF for servicing the second feeder F2 is earlier than the next scheduled time MCN for servicing the second nozzle N2; otherwise, it switches the sub-operation sequence to step S25.That is to say, Estimation Section 83 performs an estimate based on the basic idea that “a unit where the time elapsed since the last maintenance is long and the next scheduled date is approaching is highly likely to be a factor contributing to a failure due to degradation of performance”.

[0087] In the fault factor identification device 8 of the embodiment, the memory section 81 stores the operating acceptance information and the operating condition information of each of the multiple assembly operations performed in the component assembly device 1. In addition, the specification section 82 specifies the fault factor by comparing two or more fault participation rates ER when the fault occurrence rate EC in the assembly operation exceeds a predetermined value E1, and the estimation section 83 estimates the fault factor using an estimation logic that differs from that of the specification section 82.Since “specify” and “estimate” are a form of “identify”, the estimation section 83 can also identify (estimate) the error factor and reduce cases where the error factor is not identified, even in a case where the specification section 82 cannot identify (specify) the error factor. 6. Further applications and modifications of the embodiment

[0088] It should be noted that the fault factor identification device 8 of the embodiment functions in the same way in actual operating situations that differ from the assumed first and second examples, and also under conditions that differ from the assumed conditions. For example, the number of components mounted on the printed circuit board K and the number of component feed units 31 and component assembly units 45 used are generally larger than in the assumed example. Furthermore, the population for calculating the fault occurrence rate EC does not need to comprise 5000 printed circuit boards K. Additionally, the predetermined value E1 and the predetermined participation rate E2 can be set to values ​​that differ from those in the embodiment.

[0089] In a form where Estimation Section 83 operates independently of the success or failure of the specification processing of Specification Section 82, the specification result of Specification Section 82 and the estimation result of Estimation Section 83 do not always coincide. In a case where they coincide, Notification Section 84 only needs to report the matching factor of the error. In a case where they do not coincide, Notification Section 84 preferably notifies either the factor of the error reported by Specification Section 82 or both factors that do not coincide. Furthermore, the processing content of Specification Section 82's specification processing can be modified accordingly, and Estimation Section 83 must use estimation logic that differs from the modified specification processing of Specification Section 82.Furthermore, various applications and modifications can be made to the embodiments and the modified forms. List of reference symbols

[0090] 1: Component assembly device, 2: PCB transport device, 3: Component feeding device, 31: Component feeding unit, 4: Component transfer device, 43: Assembly head, 45: Component assembly unit, 47: Parts camera, 5: Control device, 52: Assembly order data, 6: Line management device, 7: PCB production line, 73: Visual PCB inspection device, 77: Feeder maintenance device, 78: Nozzle maintenance device, 8: Failure factor identification device, 81: Memory section, 82: Specification section, 83: Estimation section, 84: Notification section, LD: Log data, EC, ECF, ECN: (Failure) occurrence rate, ER: Failure participation rate, E1: Specified value, E2: Specified participation rate, MAF, MAN: Last execution time, MBF, MBN: Execution interval, MCF, MCN: Next scheduled time QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] WO 2020 / 188774

[0005]

Images