Production line analysis system, production line analysis method, learning device, and inference device

The production line analysis system addresses high data storage and server capacity costs by selectively processing device data, facilitating efficient and cost-effective identification of stoppage causes and optimizing production line operations.

JP7595512B2Active Publication Date: 2024-12-06MITSUBISHI ELECTRIC CORP
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2021068479
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-04-14
Publication Date
2024-12-06
Estimated Expiration
2041-04-14

AI Technical Summary

Technical Problem

Existing production line monitoring systems require significant data storage and server capacity, leading to high construction costs due to the need to collect and manage large amounts of data, which is inefficient and costly.

Method used

A production line analysis system that selectively collects and analyzes device data only when specific conditions are met, such as exceeding set lead and cycle times, reducing the amount of data processed and minimizing system requirements.

Benefits of technology

The system allows for cost-effective analysis of production line stoppages by reducing data processing needs, enabling efficient identification of stoppage causes and optimizing production line improvements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007595512000002
    Figure 0007595512000002
  • Figure 0007595512000003
    Figure 0007595512000003
  • Figure 0007595512000004
    Figure 0007595512000004
Patent Text Reader

Abstract

To provide a production line analysis system configured to analyze a production line with an inexpensive system configuration.SOLUTION: A production line analysis system 10A includes: a temporary acquisition unit 102 which acquires device data including operation history of production apparatuses 101 and device state thereof; a data determination unit 103 which calculates lead time of the production apparatuses 101 and cycle time of the production apparatuses 101 based on the device data, and extracts, as extraction data, device data in the case that the lead time exceeds a first set value and the cycle time exceeds a second set value, out of the above device data; a classification unit 201 which calculates downtime for each production apparatus 101 on the basis of the extraction data, and specifies a cause of the stop of a production line based on the operation history and the device state included in the extraction data; and an output unit 301 which displays the cause of the stop of the production line corresponding to the extraction data in a specific form.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to a production line analysis system, a production line analysis method, a learning device, and an inference device that analyze causes of stoppages of a production line. [Background technology]

[0002] In production lines where production equipment is installed, it is common to analyze or monitor the operation status of the production line in order to improve the output of the production line. For example, there is a method for identifying factors that are reducing the output of the production line by calculating and comparing various indices such as the operation rate and overall equipment effectiveness for each production device based on the operation status data collected from each production device.

[0003] The monitoring device in Patent Document 1 collects the time of normal operation, abnormal stop, manual stop, and stop due to the work status of each production device, and calculates the stop time caused by its own production device and the stop time caused by other production devices. This monitoring device calculates the individual operation rate of each production device by dividing the operation time by the load time, and the time allocated to production is defined as the load time, and the time obtained by subtracting the stop time caused by its own production device from the load time is defined as the operation time. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2008-112209 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology of Patent Document 1 requires the collection and storage of a large amount of information from each production device. Therefore, the technology of Patent Document 1 requires a system configuration that does not cause data loss during collection and requires sufficient server capacity for the amount of data to be collected, resulting in a problem of high system construction costs.

[0006] The present disclosure has been made in consideration of the above, and has an object to provide a production line analysis system that can analyze a production line with an inexpensive system configuration. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems and achieve the object, the present disclosure provides a production line analysis system for analyzing stoppage factors of a production line including a plurality of production devices, comprising an acquisition unit for acquiring device data including operation history and device status of each production device from each production device. The production line analysis system also comprises a data determination unit for calculating a lead time and a cycle time for each production device based on the device data, and extracting device data from the device data when the lead time exceeds a first set value and the cycle time exceeds a second set value as first extracted data, and a classification unit for calculating a downtime for each production device based on the first extracted data, and identifying a stoppage factor of the production line based on the operation history and device status included in the first extracted data. The production line analysis system also comprises a compilation unit for outputting an instruction to display the stoppage factor of the production line corresponding to the first extracted data in a specific format, and a display unit for displaying the stoppage factor of the production line corresponding to the first extracted data in a specific format according to an instruction from the compilation unit. Effect of the Invention

[0008] The production line analysis system according to the present disclosure has an effect of being able to analyze a production line with an inexpensive system configuration. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing a configuration of a production system having a production line analysis system according to a first embodiment. [Diagram 2] 1 is a flowchart showing a procedure for determining whether or not device data is necessary, performed by the production line analysis system according to the first embodiment. [Diagram 3] 1 is a flowchart showing a process for classifying causes of line stoppage by the production line analysis system according to the first embodiment; [Figure 4] FIG. 1 is a diagram for explaining a downtime calculated by the production line analysis system according to the first embodiment; [Diagram 5] 1 is a flowchart showing a procedure for outputting a cause of a line stoppage by the production line analysis system according to the first embodiment. [Figure 6] FIG. 13 is a diagram showing a first example of an analysis result of a line stoppage cause displayed by the production line analysis system according to the first embodiment. [Figure 7] FIG. 11 is a diagram showing a second example of an analysis result of a line stoppage cause displayed by the production line analysis system according to the first embodiment. [Figure 8] FIG. 1 is a diagram showing a configuration of a production system having a production line analysis system according to a second embodiment. [Figure 9] 11 is a flowchart showing a procedure for determining whether or not device data is necessary, performed by the production line analysis system according to the third embodiment. [Figure 10] 11 is a flowchart showing a processing procedure for real-time display processing of a line stoppage status by the production line analysis system according to the third embodiment. [Figure 11] FIG. 13 is a diagram showing a configuration of a learning device according to a fourth embodiment. [Figure 12] 13 is a flowchart showing a procedure of a learning process performed by the learning device according to the fourth embodiment. [Figure 13] FIG. 13 is a diagram showing a configuration of an inference device according to a fourth embodiment. [Figure 14] 13 is a flowchart showing a procedure of an inference process performed by the inference device according to the fourth embodiment. [Figure 15]FIG. 13 is a diagram showing a configuration example of a processing circuit included in a learning device according to a fourth embodiment when the processing circuit is realized by a processor and a memory. [Figure 16] FIG. 13 is a diagram showing an example of a processing circuit included in a learning device according to a fourth embodiment when the processing circuit is configured with dedicated hardware. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of a production line analysis system, a production line analysis method, a learning device, and an inference device according to the present disclosure will be described in detail with reference to the accompanying drawings.

[0011] Embodiment 1 1 is a diagram showing a configuration of a production system having a production line analysis system according to Embodiment 1. The production system 1A has a production line 20 and a production line analysis system 10A.

[0012] The production line analysis system 10A is a system that analyzes the causes of stoppage of a production line 20 in which a plurality of production devices 101 are connected. The production line analysis system 10A collects equipment data, which is information on the operation of the production devices 101, from the production devices 101 that constitute the production line 20, and analyzes the causes of stoppage of the production line 20 based on the equipment data.

[0013] The production line 20 is an example of a production line in which workpieces are processed sequentially by a plurality of linked production devices 101. Since the production devices 101 are linked, if even one of the production devices 101 stops, the production capacity of the production line 20 decreases.

[0014] Although FIG. 1 shows a case where the production devices 101 are lined up one by one in a row, the production devices 101 may be arranged so as to branch off along the way. For example, the production line 20 may include a configuration in which three (N+1)th production devices 101 are arranged after the Nth (N is a natural number) production device 101, and the Nth production device 101 is connected to each of the (N+1)th production devices 101. Note that the connection in the first embodiment is not limited to a case where the production devices 101 are physically connected to each other by a free-flow conveyor or the like, but also includes a case where the transport of workpieces between the production devices 101 is systematically connected by a self-propelled robot, an AGV (Automatic Guided Vehicle), manpower, or the like.

[0015] The production line analysis system 10A includes a network 104, a classification unit 201, a compilation unit 202, an output unit 301, a plurality of provisional acquisition units 102, and a plurality of data determination units 103.

[0016] The production line 20 has a plurality of production devices 101. One provisional acquisition unit 102 is connected to one production device 101, and one data judgment unit 103 is connected to one provisional acquisition unit 102. FIG. 1 shows a case where the production line 20 has four production devices 101, and the production line analysis system 10A is equipped with four provisional acquisition units 102 and four data judgment units 103.

[0017] The provisional acquisition unit 102 may be provided in the production equipment 101, or may be a device configured separately from the production equipment 101. The data determination unit 103 may be provided in the production equipment 101, or may be a device configured separately from the production equipment 101.

[0018] Each data judgment unit 103 is connected to a classification unit 201 via a network 104. In addition, the classification unit 201 is connected to a counting unit 202, and the counting unit 202 is connected to an output unit 301.

[0019] The production device 101 is any of various production devices that process a workpiece. For example, the production device 101 is a soldering device, a coating device, a case assembly device, a screw tightening device, an image inspection device, etc.

[0020] Each production device 101 manages time information, which is information about time. For example, the production device 101 holds, as the time information, time data indicating the time when processing of a workpiece is started in the production device 101 and the time when processing of the workpiece is completed. The time when processing of the workpiece is started, which is included in the time data, is information indicating the date and time when processing of the workpiece is started. The time when processing of the workpiece is completed, which is included in the time data, is information indicating the date and time when processing of the workpiece is completed. Note that the time from the time when processing of the first workpiece is completed to the time when processing of the second workpiece is started is the time for waiting for processing of the second workpiece to begin.

[0021] Furthermore, the production equipment 101 individually holds various types of information as equipment data, such as the ON / OFF state of a sensor, the measured value of the sensor, the position of the drive shaft, the torque of the motor, the captured image, the display color of the signal tower, the reading result of a 2D (Dimensions) code reader, and an equipment status number (described later). An example of the reading result of the code reader is a work ID (Identification) that identifies a work. The equipment data in the first embodiment is associated with time information.

[0022] Furthermore, the production equipment 101 individually sets a lead time (L / T) and a cycle time (C / T) to determine whether or not the processing is normal. The lead time in the first embodiment is the difference in time between the time when the production equipment 101 starts processing the workpiece and the time when the processing of the workpiece is completed. The cycle time in the first embodiment is the difference in time between the time when the production equipment 101 completes processing the workpiece and the time when the processing of the workpiece that was processed immediately before is completed. Note that the lead time and cycle time can be set for each production equipment 101.

[0023] The provisional acquisition unit (acquisition unit) 102 is an equipment status provisional acquisition unit that provisionally acquires an equipment status. The provisional acquisition unit 102 acquires all equipment data held in the production equipment 101. The provisional acquisition unit 102 has a storage medium for storing the acquired equipment data. The provisional acquisition unit 102 collects equipment data associated with time information from the production equipment 101. Therefore, the provisional acquisition unit 102 can acquire information on the transition of the equipment data based on the time information, i.e., information on the change in the equipment data.

[0024] The data determination unit 103 is a usage data determination unit that determines data used to analyze a line stoppage cause, which is a cause of the stoppage of the production line 20. The line stoppage cause corresponds to a cause of the stoppage of the production equipment 101. Therefore, in the following description, the line stoppage cause may be referred to as a stoppage cause of the production equipment 101 or simply as a stoppage cause.

[0025] The data determination unit 103 reads out the device data associated with the time information from the temporary acquisition unit 102. The data determination unit 103 is a calculation device that selects device data to be used when actually analyzing the cause of the line stoppage from among the device data acquired by the temporary acquisition unit 102. The selection of device data in the first embodiment means determining which period of device data is to be used for the analysis, and all device data within the selected period is used during the analysis. The data determination unit 103 sends the selected and extracted device data to the classification unit 201 via the network 104. The device data selected and extracted by the data determination unit 103 is the first extracted data.

[0026] The network 104 is a communication network for transmitting the device data determined to be used by the data determination unit 103 to the classification unit 201 located at a higher level in the production line 20. For ease of understanding, the network 104 is shown by a solid line in Fig. 1, but the network 104 connecting the data determination unit 103 and the classification unit 201 may be a wired or wireless connection.

[0027] The classification unit 201 is a stoppage cause classification unit that classifies the causes of stoppage of the production line 20. The classification unit 201 receives, from the data determination unit 103, the device data that the data determination unit 103 has determined to be used. The classification unit 201 calculates downtime and identifies the cause of line stoppage based on the device data received from the data determination unit 103, and then associates the downtime with the identified cause of line stoppage on a one-to-one basis.

[0028] The classification unit 201 sends correspondence table information, which will be described later, in which the downtime is associated with the identified line stoppage cause, to the aggregation unit 202. Note that the correspondence table information may also include information on the time when the production line 20 stopped (stoppage occurrence time, which will be described later).

[0029] The counting unit 202 is a line stoppage cause counting unit that counts the causes of line stoppage based on the correspondence table information. The counting unit 202 counts the downtime and the line stoppage cause associated with each other by the classification unit 201.

[0030] The tallying unit 202, for example, calculates the downtime of the line stoppage cause for each production device 101. The tallying unit 202 may also rearrange the correspondence table information in descending order of the downtime of the line stoppage cause. That is, the tallying unit 202 may rearrange the line stoppage causes in descending order of the downtime. Hereinafter, information in which the line stoppage causes are rearranged in descending order of the downtime is referred to as rearranged information.

[0031] The tallying unit 202 generates, for example, rearrangement information as information to be displayed on the output unit 301. The tallying unit 202 may also generate transition information indicating transition of downtime related to a specific line stoppage cause as information to be displayed on the output unit 301. The tallying unit 202 sends information to be displayed on the output unit 301, such as the rearrangement information and the transition information, to the output unit 301.

[0032] The classification unit 201 and the counting unit 202 are realized by a processing circuit having a processor and a memory. The hardware configuration of the classification unit 201 and the counting unit 202 will be described later.

[0033] The output unit 301 is a display device, such as a monitor, a tablet, or a wearable device, that can convey information to workers at the production site. The output unit 301, which is a display unit, displays, for example, rearrangement information of downtime generated by the counting unit 202. The output unit 301 also displays transition information of downtime generated by the counting unit 202.

[0034] By the output unit 301 displaying the sorting information, the worker can grasp the line stoppage causes in order of the degree of impact on the production line 20. Furthermore, by the output unit 301 displaying the transition information, the worker can grasp the transition of downtime related to a specific line stoppage cause. This allows the worker to identify the line stoppage cause of the production line 20.

[0035] Next, a process of determining whether or not the device data is necessary, performed by the production line analysis system 10A, will be described below. Fig. 2 is a flowchart showing a process of determining whether or not the device data is necessary, performed by the production line analysis system according to the first embodiment.

[0036] While the production equipment 101 is processing the workpiece, the provisional acquisition unit 102 collects the changes in the equipment data including various information within the production equipment 101 together with time information (step S10), and stores the equipment data associated with the time data.

[0037] When the production device 101 finishes processing the workpiece (step S20), the data determination unit 103 calculates the lead time and cycle time for the current processing of the workpiece from the processing completion time of the workpiece (step S30).

[0038] Next, the data determination unit 103 determines whether or not the lead time in the current workpiece machining exceeds a set lead time, which is a standard lead time set in the production device 101 (step S40).

[0039] If the current lead time exceeds the set lead time (step S40, Yes), the data determination unit 103 determines whether the cycle time in the current workpiece machining exceeds the set cycle time, which is a standard cycle time set in the production device 101 (step S50). The set lead time is a first set value, and the set cycle time is a second set value.

[0040] If the current cycle time exceeds the set cycle time (step S50, Yes), the data determination unit 103 determines that some kind of stop occurred during the processing of the current workpiece in the production device 101. That is, if the current lead time exceeds the set lead time and the current cycle time exceeds the set cycle time, the data determination unit 103 determines that the production device 101 stopped during the processing of the current workpiece. In this case, the data determination unit 103 sends the device data within the current cycle time, that is, the device data of the temporary acquisition unit 102 in the difference time between the processing completion time of the current workpiece and the processing completion time of the previous workpiece, to the classification unit 201 via the network 104 (step S60). In this way, the data determination unit 103 sends the device data in the case where the lead time and cycle time exceed the reference values, among the device data, to the classification unit 201. After this, the production device 101 starts processing the next workpiece (step S70).

[0041] On the other hand, if the current lead time does not exceed the set lead time (step S40, No), the data determination unit 103 discards the equipment data within the current cycle time without sending it to the classification unit 201 (step S65). After that, the production equipment 101 starts processing the next workpiece (step S70).

[0042] Similarly, if the current cycle time does not exceed the set cycle time (step S50, No), the data determination unit 103 discards the device data within the current cycle time without sending it to the classification unit 201 (step S65). After that, the production device 101 starts machining the next workpiece (step S70).

[0043] In this way, the data judgment unit 103 sends only the equipment data required for analyzing the cause of the line stop to the classification unit 201. When the production equipment 101 starts processing the next workpiece, the production line analysis system 10A and the production equipment 101 return to the process of step S10 and repeat the processes of steps S10 to S70.

[0044] The production device 101 may start processing the next workpiece at any timing after the processing of the workpiece is completed, i.e., after the processing of step S20. That is, the processing of the next workpiece by the production device 101 and the processing of steps S30 to S65 by the production line analysis system 10A may be executed in parallel. Also, the data judgment unit 103 may execute either the processing of step S40 or the processing of step S50 first.

[0045] Next, a process of classifying the line stoppage causes based on the device data by the production line analysis system 10A will be described. Fig. 3 is a flowchart showing a process of classifying the line stoppage causes by the production line analysis system according to the first embodiment.

[0046] The classification unit 201 receives the device data sent from the data determination unit 103 (step S110). The classification unit 201 calculates the downtime based on the device data (step S120).

[0047] Here, the downtime calculated by the classification unit 201 will be described. Fig. 4 is a diagram for explaining the downtime calculated by the production line analysis system according to the first embodiment. The horizontal axis in Fig. 4 is the time axis. The information shown in Fig. 4 is information included in the device data that the classification unit 201 receives from the data judgment unit 103. In Fig. 4, the machining of the Mth (M is a natural number) workpiece is referred to as the machining of the previous workpiece, and the machining of the (M+1)th workpiece is referred to as the machining of the next workpiece.

[0048] 4, the time when the machining of the previous workpiece is completed is indicated as time T1, and the time when the machining of the next workpiece is completed is indicated as time T4. The time from time T1 to time T4 is the current cycle time.

[0049] The production equipment 101 waits from the time when the processing of the previous workpiece is completed (time T1) to time T2 to start processing the next workpiece. The wait for processing to start represents the stoppage time of the production equipment 101 that occurs because a workpiece does not flow from the production equipment 101 that corresponds to the previous process connected to the production equipment 101. After waiting for processing to start, when the workpiece is transported to the production equipment 101 at time T2, the production equipment 101 starts processing the next workpiece, and the processing of the next workpiece is completed at time T4. If the time from time T3 to time T4 is the set cycle time, the time exceeding that time, from time T2 to time T3, is downtime.

[0050] The classification unit 201 calculates the downtime for the current work by subtracting the set cycle time and the time required to wait for the start of processing from the cycle time for the current work.

[0051] The classification unit 201 calculates the downtime caused by the production equipment 101 by excluding the time spent waiting for the start of processing from the time that exceeds the set cycle time of the production equipment 101, and determines this calculated downtime as the downtime. Therefore, if the downtime is a negative value, the classification unit 201 discards the equipment data because the equipment has stopped due to the effect of waiting for the start of processing, and proceeds to the next process only if the downtime is a positive value.

[0052] Specifically, the classification unit 201 determines whether the downtime is a positive value (step S130). If the downtime is a positive value (step S130, Yes), the classification unit 201 analyzes the equipment data and identifies the cause of the line stoppage that caused the downtime (step S140).

[0053] Here, a description will be given of an example of a method for identifying the cause of line stoppage by the classification unit 201. Note that the method for identifying the cause of line stoppage described here is an example in the first embodiment, and a preferred method for identifying the cause of line stoppage is adopted depending on the device specifications of each production device 101.

[0054] In the first embodiment, the classification unit 201 identifies the cause of line stoppage based on the operation history of a signal tower provided in the production equipment 101 and the equipment status of the production equipment 101 when the signal tower operated. The operation history of the signal tower and the equipment status of the production equipment 101 are information included in the equipment data. The equipment status of the production equipment 101 is indicated by identification information indicating the equipment status, for example, an equipment status No. (number) which is the number of the equipment status.

[0055] In the first step, the classification unit 201 extracts the display color state of the signal tower based on the operation history of the signal tower. In this way, the classification unit 201 identifies the equipment state when the line stop factor that is the cause of the downtime occurred. For example, if the stop is due to an equipment abnormality, the color of the signal tower changes from green, which indicates normal production, to red, which indicates an abnormality, so the classification unit 201 can determine that the cause of the downtime is the equipment abnormality. In addition, if the stop is due to a call for an operator, the color of the signal tower changes from green, which indicates normal production, to yellow, so the classification unit 201 can determine that the cause of the downtime is the call for an operator, that is, a delayed response by the operator. Even if there is no change in the display color of the signal tower when a line stop occurs, the classification unit 201 can determine that the downtime occurred when the signal tower was green, which indicates normal production.

[0056] In a second step, the classification unit 201 extracts from the equipment data an equipment status number for the equipment status identified in the first step. For example, if the line is stopped due to an equipment abnormality, the equipment status number is a number set for each type of error. In addition, if a worker is called, the equipment status number is a number set for each type of material supply, material discharge, and auxiliary material replacement. If normal production is occurring, the equipment status number is the number of the cycle operation that has exceeded the specified time. In the first embodiment, the classification unit 201 classifies the line stop cause based on the equipment status number, so that the classification unit 201 can easily classify the line stop cause.

[0057] After the classification unit 201 identifies the cause of the line stop of the production equipment 101 in the above manner, the classification unit 201 associates the cause of the line stop with downtime (step S150). At this time, it is preferable that the classification unit 201 also associates the cause of the line stop with the downtime, which is the time when the cause of the line stop occurred, in order to facilitate analysis using a graph at the time of output.

[0058] In this case, the classification unit 201 associates the stop occurrence time indicating the time when the production equipment 101 stopped, the equipment status, the equipment status No., and the downtime indicating the period during which the production equipment 101 stopped. Then, the classification unit 201 generates a data table in which the associated information is arranged in a row and stored. For example, the classification unit 201 provides the first to fourth columns in the data table. The first column stores "2020 / 10 / 5 11:49:37" indicating the stop occurrence time, the second column stores "equipment abnormality" indicating the equipment status, the third column stores "11" indicating the equipment status No., and the fourth column stores "178" indicating the downtime in seconds of the stop. The classification unit 201 sends the correspondence table information, which is the data associated in this way, to the aggregation unit 202 (step S160).

[0059] If the downtime is not a positive value (step S130, No), the classification unit 201 discards the device data received from the data judgment unit 103 (step S170). After step S160 or step S170, the production line analysis system 10A returns to the process of step S110 and repeats the processes of steps S110 to S170.

[0060] Next, a process of outputting a line stoppage cause by the production line analysis system 10A will be described below. Fig. 5 is a flowchart showing a process of outputting a line stoppage cause by the production line analysis system according to the first embodiment.

[0061] The aggregation unit 202 stores the correspondence table information sent from the classification unit 201. That is, the aggregation unit 202 stores the line stop cause associated with the downtime (step S210). For example, the aggregation unit 202 stores the correspondence table information generated in a list format consisting of rows and columns.

[0062] The classification unit 201 sends the stored correspondence table information to the output unit 301 at a specific timing (step S220). The counting unit 202 sends an instruction to the output unit 301 to display the information on the line stop cause in the correspondence table information in a specific format. Note that the counting unit 202 is not limited to sending the correspondence table information to the output unit 301 as it is, and may send specific information calculated using the correspondence table information, i.e., information on the line stop cause, to the output unit 301.

[0063] The tallying unit 202, for example, accumulates the downtime for each cause of line stoppage of each production device 101, and causes the output unit 301 to display the accumulated downtime for each cause of line stoppage of each production device 101.

[0064] Furthermore, the tallying unit 202 may cause the output unit 301 to display the daily transition of the downtime corresponding to a specific line stoppage cause. In this case, the tallying unit 202 calculates the average downtime, which is the average value of the downtime of each production device 101, on a daily basis for the specific line stoppage cause. Then, the tallying unit 202 adds up the average downtime included in one day, and causes the output unit 301 to display the added up average downtime for each day.

[0065] The output unit 301 displays the information on the cause of the line stop in a specific format according to an instruction from the tallying unit 202 (step S230). After the process of step S230, the production line analysis system 10A returns to the process of step S210 and repeats the processes of steps S210 to S230.

[0066] 6 is a diagram showing a first example of an analysis result of the line stoppage cause displayed by the production line analysis system according to the embodiment 1. The graph in FIG. 6 shows the relationship between the line stoppage cause and the downtime in a specific production device 101.

[0067] The output unit 301 displays the information on the line stoppage cause as, for example, a bar graph. The horizontal axis of Fig. 6 indicates the line stoppage cause, and the vertical axis indicates the accumulated downtime, which is the accumulated downtime. X of the cause X shown in Fig. 6 (X is 33, 5, etc.) is information that identifies the line stoppage cause.

[0068] The counting unit 202 calculates the accumulated downtime by accumulating the downtime for each cause of line stoppage of each production device 101. The counting unit 202 causes the output unit 301 to display the accumulated downtime for each cause of line stoppage of each production device 101.

[0069] Here, the output unit 301 displays information on the line stoppage causes in the order of the longest downtime. That is, the output unit 301 displays the line stoppage causes that caused the capacity degradation of the production line 20 in the order of the longest downtime, i.e., the stoppage time of the production line 20.

[0070] Fig. 7 is a diagram showing a second example of the analysis result of the line stoppage cause displayed by the production line analysis system according to the first embodiment. The graph in Fig. 7 shows the daily change in the line stoppage cause. The output unit 301 displays the change in the line stoppage cause as, for example, a line graph. The horizontal axis in Fig. 7 shows the day when the production line 20 stopped, and the vertical axis shows the accumulated average downtime, which is the accumulated average downtime. Fig. 7 shows the change in cause 23, whose identification number for the line stoppage cause is "23", i.e., the trend.

[0071] The aggregation unit 202 accumulates the average downtime, which is the average value of the downtime of each production device 101 for a specific line stop cause, within a period of one day, and displays the accumulated average downtime of each production device 101 for one day on the output unit 301.

[0072] As described above, according to the first embodiment, the production line analysis system 10A uses only the device data when the set lead time and the set cycle time are exceeded for analysis, so there is no need to analyze all data during production. This reduces the network load and the total amount of data in the production line analysis system 10A, so that the cost required to build the production line analysis system 10A can be suppressed. Therefore, the production line analysis system 10A can analyze the production line 20 with an inexpensive system configuration.

[0073] Furthermore, since the production line analysis system 10A associates downtime, which is the time the production line 20 is stopped, with the line stoppage factors of the production line 20, it is possible to compare the stoppage factors of each production device 101 in order of the degree of influence on the production line 20. This allows the operator to proceed with improvement activities for the production line 20 in the correct order.

[0074] Embodiment 2 Next, a second embodiment will be described with reference to Fig. 8. In the second embodiment, one classification unit 201 is connected to one data judgment unit 103, and each classification unit 201 is connected to a counting unit 202 via a network 104.

[0075] Fig. 8 is a diagram showing a configuration of a production system having a production line analysis system according to embodiment 2. Among the components in Fig. 8, components that achieve the same functions as those in production system 1A of embodiment 1 shown in Fig. 1 are given the same reference numerals, and duplicated explanations will be omitted.

[0076] The production system 1B has a production line analysis system 10B and a production line 20. The production line analysis system 10B is connected to the production line 20 in the same manner as the production line analysis system 10A. The production line analysis system 10B has a network 104, a plurality of classification units 201, a counting unit 202, an output unit 301, a plurality of temporary acquisition units 102, and a plurality of data judgment units 103. Thus, the production line analysis system 10B has a plurality of classification units 201, compared to the production line analysis system 10A. Note that the classification units 201 may be included in the production equipment 101, or may be a device configured separately from the production equipment 101.

[0077] In the production line analysis system 10B, one classification unit 201 is connected to one data judgment unit 103. Each classification unit 201 is connected to a network 104. In addition, a counting unit 202 is connected to the network 104.

[0078] The production line analysis system 10B can identify the cause of the line stoppage of the production line 20 in a similar procedure to that of the production line analysis system 10A. In the production line analysis system 10B, the information sent to the counting unit 202 via the network 104 is limited to the identified cause of the line stoppage, downtime, and the like.

[0079] As described above, according to the second embodiment, the information sent to the counting unit 202 via the network 104 is limited to some numerical data, so that the production line analysis system 10B can be stabilized. In addition, since the analysis of the cause of the stoppage of the production line 20 can be executed on the production equipment 101 side, the production line analysis system 10B can execute even complicated analysis processing that takes into account the change over time of production facilities such as the production equipment 101. This allows the production line analysis system 10B to identify the cause of the stoppage of the production line 20 in detail.

[0080] Embodiment 3 Next, a third embodiment will be described with reference to Figures 9 and 10. In the third embodiment, even if a workpiece is being processed, the cause of the line stoppage is analyzed and identified when the set lead time and set cycle time are exceeded.

[0081] The production line analysis system 10A of the embodiment 3 has the same configuration as the production line analysis system 10A of the embodiment 1. The production line analysis system 10A of the embodiment 3 is different from the production line analysis system 10A of the embodiment 1 in the processing procedure of the process for determining whether or not device data is required.

[0082] Fig. 9 is a flowchart showing a process procedure of a process of determining whether or not device data is necessary, performed by the production line analysis system according to the embodiment 3. Note that, among the processes in Fig. 9, the description of the same processes as those described in Fig. 2 will be omitted.

[0083] While the production equipment 101 is processing the workpiece, the provisional acquisition unit 102 collects the changes in the equipment data including various information within the production equipment 101 together with time information (step S310), and stores the equipment data associated with the time data.

[0084] While the production device 101 is machining the workpiece, the data judgment unit 103 calculates the expected values ​​of the lead time and cycle time for machining the current workpiece (step S320). The expected value of the lead time is the difference in time between the current time and the start time of machining the workpiece, and the expected value of the cycle time is the difference in time between the current time and the completion time of the workpiece that was machined immediately before. The data judgment unit 103 constantly calculates the expected values ​​while the workpiece is being machined.

[0085] The data determination unit 103 determines whether or not the expected value of the lead time for the current workpiece machining exceeds the set lead time (step S330).

[0086] If the expected value of the current lead time does not exceed the set lead time (step S330, No), the data determination unit 103 returns to the process of step S320.

[0087] If the estimated value of the current lead time exceeds the set lead time (step S330, Yes), the data judgment unit 103 judges whether the estimated value of the cycle time for the current workpiece processing exceeds the set cycle time (step S340).

[0088] If the expected value of the current cycle time does not exceed the set cycle time (step S340, No), the data determination unit 103 returns to the process of step S320.

[0089] If the expected value of the current cycle time exceeds the set cycle time (step S340, Yes), the data determination unit 103 determines that the production equipment 101 has stopped during the processing of the current workpiece. That is, if the expected value of the current lead time exceeds the set lead time and the expected value of the current cycle time exceeds the set cycle time, the data determination unit 103 determines that the production equipment 101 has stopped during the processing of the current workpiece. In this case, the data determination unit 103 sends the equipment data of the provisional acquisition unit 102 at the time when the set lead time and the set cycle time are exceeded to the classification unit 201 (step S350). That is, the data determination unit 103 selects and extracts the equipment data at the time when the set lead time and the set cycle time are exceeded, and sends the extracted equipment data to the classification unit 201 via the network 104. The equipment data selected and extracted by the data determination unit 103 is the second extracted data.

[0090] After that, when the production equipment 101 finishes processing the current workpiece (step S360), the data judgment unit 103 sends the lead time and cycle time to the classification unit 201 (step S370). After that, the production equipment 101 starts processing the next workpiece (step S380). The production line analysis system 10A and the production equipment 101 return to the process of step S310 and repeat the processes from step S310 to S380.

[0091] The production equipment 101 may start machining the next workpiece at any timing after step S360. That is, the production equipment 101 may process the next workpiece and the data determination unit 103 may perform the process of step S370 in parallel. In this way, in the third embodiment, even during machining of a workpiece, the data determination unit 103 sends the equipment data at the time when the set lead time and the set cycle time have exceeded to the classification unit 201.

[0092] 10 is a flowchart showing a processing procedure of a real-time display process of a line stop status by the production line analysis system according to the embodiment 3. The line stop status described in the embodiment 3 is information indicating whether the state of the production equipment 101 is an “abnormal” state or a “normal” state, that is, status information.

[0093] The classification unit 201 receives the equipment data selected by the data determination unit 103 based on the expected value from the data determination unit 103 (step S410). The classification unit 201 analyzes the equipment data and identifies the cause of the line stoppage that caused the downtime (step S420).

[0094] The classification unit 201 acquires the average downtime of the line stoppage cause from the counting unit 202 (step S430). The classification unit 201 sends the identified line stoppage cause and the average downtime of the line stoppage cause to the counting unit 202 in association with each other.

[0095] The counting unit 202 causes the output unit 301 to output in real time information indicating that the current state of the production equipment 101 is an "abnormal" state. The counting unit 202 also causes the output unit 301 to output in real time the line stoppage cause and the average downtime of the line stoppage cause received from the classification unit 201 as the line stoppage cause and average downtime of the "abnormal" state. That is, the counting unit 202 causes the output unit 301 to display in real time the "abnormal" state of the production equipment 101, the line stoppage cause, and the average downtime (step S440).

[0096] The tallying unit 202 causes the output unit 301 to display information indicating that the current state of the production equipment 101 is "normal" at the time when the processing of the workpiece is completed and the downtime is determined (step S450). That is, the tallying unit 202 changes the information indicating the current state of the production equipment 101 from "abnormal" to "normal". In this case, as in the first embodiment, the production line analysis system 10A causes the output unit 301 to display the determined downtime and line stop cause as the tallying results, following the processing procedures of the flowcharts described in FIG. 3 and FIG. 5.

[0097] Thereafter, the production line analysis system 10A and the production device 101 return to the process of step S410 and repeat the processes of steps S410 to S450. Note that the production line analysis system 10B may execute the processes described with reference to FIGS.

[0098] As described above, in the third embodiment, the production line analysis system 10A causes the output unit 301 to display in real time the line stoppage cause, the stoppage status of the production equipment 101, and the average downtime, which is the expected stoppage time of the production equipment 101, without waiting for the completion of processing of the workpiece. This allows the worker to grasp the line stoppage cause, the stoppage status of the production equipment 101, and the average downtime, and therefore the order of recovery work for the production equipment 101 at the production site can be optimized.

[0099] Embodiment 4 Next, a fourth embodiment will be described with reference to Fig. 11 to Fig. 16. In the fourth embodiment, the recovery priority is learned by machine learning when the stop status of the production equipment 101 is displayed in real time as described in the third embodiment. The recovery priority is information indicating the priority of the process for an operator to recover the production equipment 101.

[0100] <Learning Phase> 11 is a diagram showing a configuration of a learning device according to the embodiment 4. The learning device 50 is a computer that learns the recovery priority when the production line analysis system 10A described in the embodiment 3 is in an “abnormal” state.

[0101] The learning device 50 includes a data acquisition unit 51 and a model generation unit 52. The data acquisition unit 51 acquires behavioral data and status data as learning data. The data acquisition unit 51 acquires behavioral data from workers and the like, and acquires status data from the production line analysis system 10A.

[0102] The behavioral data acquired by the data acquiring unit 51 is the recovery priority of the production equipment 101. The status data acquired by the data acquiring unit 51 is the average cycle time of each production equipment 101, the stoppage status of each production equipment 101, the average downtime for each line stoppage cause of each production equipment 101, and the amount of work in progress between each production equipment 101.

[0103] The stop status of each production equipment 101 includes not only the equipment status, whether the production equipment 101 is in a "normal" or "abnormal" state, but also time information, such as how many seconds have passed since the set lead time and set cycle time were exceeded. When the equipment status is in an "abnormal" state, the cause of the stop is indicated by the color displayed on the signal tower, the equipment status number, etc.

[0104] The model generation unit 52 learns the recovery priority of the production equipment 101 in the input state based on learning data including the recovery priority of the production equipment 101, the average cycle time of each production equipment 101, the stop status of each production equipment 101, the average downtime for each stop cause of each production equipment 101, and the amount of work in progress between each production equipment 101. That is, the model generation unit 52 generates a learned model that infers the recovery priority of the production equipment 101 when the average cycle time of each production equipment 101, the stop status of each production equipment 101, the average downtime for each stop cause of each production equipment 101, and the amount of work in progress between each production equipment 101 in the production line analysis system 10A of the production line 20 are input. The recovery priority of the production equipment 101 in the input state is an appropriate recovery priority of the production equipment 101 corresponding to the input state data.

[0105] The learning algorithm used by the model generation unit 52 may be a known algorithm such as supervised learning, unsupervised learning, or reinforcement learning. As an example, a case where the model generation unit 52 applies reinforcement learning to the learning algorithm will be described. In reinforcement learning, an agent (acting subject) in a certain environment observes the current state (environmental parameters) and determines the action to be taken. The environment changes dynamically due to the agent's actions, and the agent is given a reward according to the change in the environment. The agent repeats this process and learns the action policy that will obtain the most reward through a series of actions. Q-learning and TD-learning are known as representative methods of reinforcement learning. For example, in the case of Q-learning, a general update formula for the action value function Q(s, a) is expressed by the following formula (1).

[0106]

number

[0107] In formula (1), s t represents the state of the environment at time t, and a t represents the action at time t. Action a t Therefore, the state is s t Changes to +1. t+1 represents the reward that can be obtained depending on the change in state, γ represents the discount rate, and α represents the learning coefficient. Note that γ is in the range of 0<γ≦1, and α is in the range of 0<α≦1. If the recovery priority of the production equipment 101 is action a t The average cycle time of each production device 101, the stoppage status of each production device 101, the average downtime for each stoppage cause of each production device 101, and the amount of work in progress between each production device 101 are in state s t At time t, the state s t Best Practices in a t Learn.

[0108] The update formula expressed by formula (1) increases the action value Q if the action value Q of the action a with the highest Q value at time t+1 is greater than the action value Q of the action a executed at time t, and decreases the action value Q in the opposite case. In other words, the update formula expressed by formula (1) updates the action value function Q(s, a) so that the action value Q of the action a at time t approaches the best action value at time t+1. As a result, the best action value in a certain environment is propagated sequentially to the action value in the previous environment.

[0109] As described above, when generating a trained model by reinforcement learning, the model generation unit 52 includes a reward calculation unit 53 and a function update unit 54.

[0110] The remuneration calculation unit 53 calculates the remuneration based on the recovery priority of the production equipment 101, the average cycle time of each production equipment 101, the stoppage status of each production equipment 101, the average downtime for each stoppage cause of each production equipment 101, and the amount of work in progress between each production equipment 101. The remuneration calculation unit 53 calculates the remuneration r based on the output of the production line 20. For example, when the output of the production line 20 increases, the remuneration r is increased (for example, a remuneration of "1" is given), whereas when the output of the production line 20 decreases, the remuneration r is decreased (for example, a remuneration of "-1" is given). Note that the output of the production line 20 can be freely determined by adopting the output of the production equipment 101 that is the final process of the production line 20, or by adopting the output of the production equipment 101 that is the bottleneck process of the production line 20. In other words, the output of which production equipment 101 is adopted as the output of the production line 20 may be appropriately selected according to the operation mode of each production site.

[0111] The function update unit 54 updates the function for determining the recovery priority of the production equipment 101 in the input state in accordance with the reward calculated by the reward calculation unit 53, and outputs the function to the learned model storage unit 70 as a learned model. For example, in the case of Q-learning, the function update unit 54 updates the action value function Q(s t ,a t) is used as a function for calculating the recovery priority of the production equipment 101 in the input state.

[0112] The model generation unit 52 repeatedly executes the above-mentioned learning. The learned model storage unit 70 stores the action-value function Q(s t ,a t ), i.e., memorize the trained model.

[0113] Next, a process in which the learning device 50 learns the recovery priority will be described with reference to Fig. 12. Fig. 12 is a flowchart showing the procedure of the learning process by the learning device according to the fourth embodiment.

[0114] The data acquisition unit 51 acquires, as learning data, the recovery priority of the production equipment 101, the average cycle time of each production equipment 101, the stoppage status of each production equipment 101, the average downtime for each stoppage cause of each production equipment 101, and the amount of work in progress between each production equipment 101 (step S510).

[0115] The model generation unit 52 calculates the remuneration based on the recovery priority of the production equipment 101, the average cycle time of each production equipment 101, the stoppage status of each production equipment 101, the average downtime for each stoppage cause of each production equipment 101, and the amount of work in progress between each production equipment 101 (step S520). Specifically, the remuneration calculation unit 53 acquires the recovery priority of the production equipment 101, the average cycle time of each production equipment 101, the stoppage status of each production equipment 101, the average downtime for each stoppage cause of each production equipment 101, and the amount of work in progress between each production equipment 101, and determines whether to increase or decrease the remuneration based on the predetermined output of the production line 20.

[0116] When it is determined that the remuneration should be increased (step S520, the output of the production line 20 increases), the remuneration calculation unit 53 increases the remuneration (step S530). That is, when the output of the production line 20 increases and thus satisfies the remuneration increase criterion, the remuneration calculation unit 53 increases the remuneration.

[0117] On the other hand, when it is determined that the remuneration should be decreased (step S520, the output of the production line 20 is decreased), the remuneration calculation unit 53 decreases the remuneration (step S540). That is, when the output of the production line 20 is decreased and the remuneration decrease criterion is satisfied, the remuneration calculation unit 53 decreases the remuneration.

[0118] The function update unit 54 updates the action value function Q(s t ,a t ) is updated (step S550).

[0119] The learning device 50 repeatedly executes the above steps S510 to S550, and generates the action-value function Q(s t ,a t ) is stored as a learned model in the learned model storage unit 70.

[0120] The learning device 50 in embodiment 4 has been described as storing the learned model in a learned model storage unit 70 provided outside the learning device 50, but the learned model storage unit 70 may also be located inside the learning device 50.

[0121] <Utilization phase> 13 is a diagram showing a configuration of an inference device according to the embodiment 4. The inference device 60 is a computer that infers the recovery priority when the production line analysis system 10A described in the embodiment 3 is in an “abnormal” state.

[0122] The inference device 60 includes a data acquisition unit 61 and an inference unit 62. The data acquisition unit 61 acquires the average cycle time of each production device 101, the stoppage status of each production device 101, the average downtime for each stoppage cause of each production device 101, and the amount of work in progress between each production device 101.

[0123] The inference unit 62 infers the recovery priority of the production equipment 101 in the input state, using the learned model stored in the learned model storage unit 70. That is, the inference unit 62 inputs the average cycle time of each production equipment 101 acquired by the data acquisition unit 61, the stoppage status of each production equipment 101, the average downtime for each stoppage cause of each production equipment 101, and the amount of work in progress between each production equipment 101, into this learned model, and thereby can infer the recovery priority of the production equipment 101 that is suitable for these inputs.

[0124] In the fourth embodiment, the inference device 60 has been described as outputting the recovery priority of the production equipment 101 using a trained model trained by the model generation unit 52 for the production line analysis system 10A connected to the production line 20, but the inference device 60 may acquire a trained model trained for another production line analysis system connected to another production line. In this case, the inference device 60 outputs the recovery priority of the production equipment 101 using a trained model trained for another production line analysis system connected to another production line.

[0125] Next, a process in which the inference device 60 infers the recovery priority will be described with reference to Fig. 14. Fig. 14 is a flowchart showing the procedure of the inference process by the inference device according to the fourth embodiment.

[0126] The data acquisition unit 61 acquires the average cycle time of each production device 101, the stoppage status of each production device 101, the average downtime for each stoppage cause of each production device 101, and the amount of work in progress between each production device 101 as data for inference (step S610).

[0127] The inference unit 62 inputs the average cycle time of each production equipment 101, the stoppage status of each production equipment 101, the average downtime for each stoppage cause of each production equipment 101, and the amount of work in progress between each production equipment 101 to the learned model stored in the learned model storage unit 70 (step S620), and obtains a recovery priority corresponding to the input information. The inference unit 62 outputs the obtained recovery priority to the output unit 301 of the production line analysis system 10A (step S630).

[0128] The output unit 301 of the production line analysis system 10A displays the recovery priority level sent from the inference unit 62, thereby notifying the worker at the production site of the recovery priority level of the production equipment 101.

[0129] Thereafter, the worker at the production site performs recovery work for the production equipment 101 while checking the recovery priority of the production equipment 101 displayed on the output unit 301 (step S640).

[0130] By applying the above-described learning device 50 and inference device 60 to the production line analysis system 10A, the inference device 60 can automatically present a recovery priority for increasing the output of the production line 20 according to the stoppage status of each production device 101. Then, by having a worker at the production site perform recovery work for the production device 101 according to the recovery priority of the production device 101, it is possible to suppress a decrease in the output of the production line 20 caused by downtime due to each line stoppage cause.

[0131] In the fourth embodiment, a case has been described in which reinforcement learning is applied to the learning algorithm used by the inference unit 62, but the learning algorithm is not limited to reinforcement learning. As for the learning algorithm used by the inference unit 62, supervised learning, unsupervised learning, semi-supervised learning, or the like can also be applied in addition to reinforcement learning.

[0132] Deep learning, which learns to extract features themselves, may also be used as a learning algorithm used in the model generation unit 52. The model generation unit 52 may also perform machine learning according to other known methods, such as a neural network, genetic programming, functional logic programming, or a support vector machine.

[0133] The learning device 50 and the inference device 60 may be devices separate from the production line analysis system 10A and connected to the production line analysis system 10A via a network such as the network 104. The learning device 50 and the inference device 60 may be built into the production line analysis system 10A. Furthermore, the learning device 50 and the inference device 60 may exist on a cloud server.

[0134] The model generating unit 52 may learn the recovery priority of the production equipment 101 in the input state by using learning data acquired from a plurality of production line analysis systems 10A. The model generating unit 52 may acquire learning data from a plurality of production line analysis systems 10A used in the same area, or may learn the recovery priority of the input production equipment 101 by using learning data collected from a plurality of production line analysis systems 10A operating independently in different areas. The production line analysis system 10A that collects the learning data may be added to the target midway or removed from the target. Furthermore, the learning device 50 that has learned the recovery priority for a certain production line analysis system 10A may be applied to another production line analysis system 10A, and the recovery priority for the other production line analysis system 10A may be re-learned and updated. The learning device 50 and the inference device 60 may be applied to the production line analysis system 10B.

[0135] Here, we will explain the hardware configurations of the learning device 50 and the inference device 60. Since the learning device 50 and the inference device 60 have similar hardware configurations, here, the hardware configuration of the learning device 50 will be explained.

[0136] The learning device 50 is realized by a processing circuit. The processing circuit may be a processor and memory that executes a program stored in a memory, or may be dedicated hardware such as a dedicated circuit. The processing circuit is also called a control circuit.

[0137] FIG. 15 is a diagram showing an example of the configuration of a processing circuit in the learning device according to the fourth embodiment, in the case where the processing circuit is realized by a processor and a memory. The processing circuit 90 shown in FIG. 15 is a control circuit, and includes a processor 91 and a memory 92. When the processing circuit 90 is configured with the processor 91 and the memory 92, each function of the processing circuit 90 is realized by software, firmware, or a combination of software and firmware. The software or firmware is described as a program and stored in the memory 92. In the processing circuit 90, each function is realized by the processor 91 reading and executing the program stored in the memory 92. That is, the processing circuit 90 includes a memory 92 for storing a program that results in the processing of the learning device 50 being executed. This program can also be said to be a program for causing the learning device 50 to execute each function realized by the processing circuit 90. This program may be provided by a storage medium in which the program is stored, or may be provided by other means such as a communication medium. The above program can also be said to be a program for causing the learning device 50 to execute the learning process.

[0138] Here, the processor 91 is, for example, a CPU (Central Processing Unit), a processing device, an arithmetic unit, a microprocessor, a microcomputer, or a DSP (Digital Signal Processor). The processor 91 is included in a PC (Personal Computer) or a PLC (Programmable Logic Controller). A PLC is also called a sequencer.

[0139] The memory 92 may be, for example, a non-volatile or volatile semiconductor memory such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable ROM), an EEPROM (registered trademark) (Electrically EPROM), a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, or a DVD (Digital Versatile Disc).

[0140] FIG. 16 is a diagram showing an example of a processing circuit provided in the learning device according to the fourth embodiment when the processing circuit is configured with dedicated hardware. The processing circuit 93 shown in FIG. 16 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination of these. The processing circuit 93 may be partially realized by dedicated hardware and partially realized by software or firmware. In this way, the processing circuit 93 can realize each of the above-mentioned functions by dedicated hardware, software, firmware, or a combination of these.

[0141] The provisional acquisition unit 102 and the data determination unit 103 also have the hardware configurations described in Fig. 15 and Fig. 16. The classification unit 201 and the counting unit 202 also have the hardware configurations described in Fig. 15 and Fig. 16.

[0142] The configurations shown in the above embodiments are merely examples, and may be combined with other known technologies, or the embodiments may be combined with each other. Also, parts of the configurations may be omitted or modified without departing from the spirit of the invention. [Explanation of symbols]

[0143] 1A, 1B Production system, 10A, 10B Production line analysis system, 20 Production line, 50 Learning device, 51 Data acquisition unit, 52 Model generation unit, 53 Reward calculation unit, 54 Function update unit, 60 Inference device, 61 Data acquisition unit, 62 Inference unit, 70 Learned model storage unit, 90, 93 Processing circuit, 91 Processor, 92 Memory, 101 Production device, 102 Temporary acquisition unit, 103 Data judgment unit, 104 Network, 201 Classification unit, 202 Aggregation unit, 301 Output unit.

Claims

1. A production line analysis system for analyzing causes of stoppage of a production line including a plurality of production devices, an acquisition unit that acquires device data including an operation history of each of the production devices and a device status of each of the production devices from each of the production devices; a data determination unit that calculates a lead time and a cycle time for each of the production devices based on the device data, and extracts, from the device data, device data in which the lead time exceeds a first set value and the cycle time exceeds a second set value, as first extracted data; a classification unit that calculates downtime for each of the production devices based on the first extracted data, and identifies a cause of a stoppage of the production line based on the operation history and the device status included in the first extracted data; a compilation unit that outputs an instruction to display the cause of the stoppage of the production line corresponding to the first extracted data in a specific format; a display unit that displays the cause of the stoppage of the production line corresponding to the first extracted data in a specific format in accordance with an instruction from the tabulation unit; Equipped with A production line analysis system comprising:

2. The counting unit accumulates the downtime for each of the stoppage causes, The display unit displays the accumulated downtime for each of the stop causes in a graph.

2. The production line analysis system according to claim 1.

3. the counting unit calculates an average value of the downtime for each of the production devices for the specific stoppage cause, and accumulates the calculated average value of the downtime within a period of one day; The display unit displays the accumulated average value of the downtime in a graph.

2. The production line analysis system according to claim 1.

4. The equipment data includes a display color of a signal tower disposed on each of the production equipments and identification information for identifying the equipment status, The classification unit identifies a cause of the stoppage of the production line based on the display color of the signal tower and the identification information.

4. The production line analysis system according to claim 1, wherein the production line analysis system comprises:

5. the data determination unit and the classification unit are connected via a network, The data determination unit transmits the first extracted data to the classification unit via the network.

5. The production line analysis system according to claim 1, wherein the production line analysis system comprises:

6. The pre-classification unit and the counting unit are connected by a network, The classification unit transmits the cause of the stoppage of the production line to the counting unit via the network.

5. The production line analysis system according to claim 1, wherein the production line analysis system comprises:

7. the data judgment unit calculates an expected value of a lead time for each of the production devices and an expected value of a cycle time for each of the production devices based on the device data while the production devices are processing the workpiece, and extracts, from the device data, device data when the expected value of the lead time exceeds the first set value and the expected value of the cycle time exceeds the second set value as second extracted data; the classification unit calculates downtime for each of the production devices based on the second extracted data, and identifies a cause of the stoppage of the production line based on the operation history and the device status included in the second extracted data; the tabulation unit outputs an instruction to display in real time in a specific format a cause of stoppage of the production line corresponding to the second extracted data; the display unit displays in real time, in a specific format, the cause of the stoppage of the production line corresponding to the second extracted data, in accordance with an instruction from the aggregation unit.

7. The production line analysis system according to claim 1, wherein the production line analysis system comprises:

8. the tallying unit outputs an instruction to display the average value of the downtime and the status of the production equipment in real time; The display unit displays the average value of the downtime and the status of the production equipment in real time.

8. The production line analysis system according to claim 7.

9. A production line analysis method for analyzing stoppage causes of a production line including a plurality of production devices, comprising the steps of: an acquisition step of acquiring device data including an operation history of each of the production devices and a device status of each of the production devices from each of the production devices; a data determination step of calculating a lead time and a cycle time for each of the production devices based on the device data, and extracting, from the device data, device data in which the lead time exceeds a first set value and the cycle time exceeds a second set value, as first extracted data; a classification step of calculating downtime for each of the production devices based on the first extracted data, and identifying a cause of a stoppage of the production line based on the operation history and the device status included in the first extracted data; a compilation step of outputting an instruction to display the cause of the stoppage of the production line corresponding to the first extracted data in a specific format; a display step of displaying a cause of a stop of the production line corresponding to the first extracted data in a specific format; Including, A production line analysis method comprising:

10. a data acquisition unit that acquires learning data from a production line analysis system that analyzes factors that cause a production line including a plurality of production devices, the learning data including an average cycle time that is an average value of cycle times of each of the production devices, status information that indicates whether each of the production devices is in a normal state or an abnormal state, an average downtime that is an average value of downtime for each factor that causes a production line to be stopped, a work in progress amount between the production devices, and a recovery priority of the production devices in relation to the average cycle time, the status information, the average downtime, and the work in progress amount between the production devices; a model generation unit that generates a trained model for inferring the recovery priority from the average cycle time, the state information, the average downtime, and the amount of work in progress between the production devices, using the learning data; and Equipped with A learning device characterized by:

11. a data acquisition unit that acquires, from a production line analysis system that analyzes factors that cause a production line including a plurality of production devices, an average cycle time that is an average value of the cycle times of the production devices, status information that indicates whether the production devices are in a normal state or an abnormal state, an average downtime that is an average value of downtime for each factor that causes a production line to be stopped, and an amount of work in progress between the production devices; an inference unit that uses a trained model for inferring a recovery priority of the production device in terms of the average cycle time, the status information, the average downtime, and the amount of work in progress between the production devices, and outputs the recovery priority from the average cycle time, the status information, the average downtime, and the amount of work in progress between the production devices, which are acquired by the data acquisition unit; Equipped with An inference device comprising:

Citation Information

Patent Citations

  • Data totaling processor and recording medium recoding program for data totaling processing

    JP2000123085A

  • Operating condition analyzing device for production line, its method, operating condition analyzing program and operating condition analyzing system

    JP2004280324A

  • Lead time managing method and lead time managing program

    JP2004326481A

  • Data collecting unit, display system, and tool

    JP2005128721A

  • System for managing production equipment

    JP2006113965A