Production system, production line analysis method, learning device, inference device, learned model, and method for generating learned model

Abstract

The present invention comprises: a key performance indicator calculation unit that calculates, on the basis of device data related to a production history of each production device (101), a value of a key performance indicator which is a criterion for determining whether the operation state of a production line (20) is acceptable; and a stop factor calculation unit (302) that calculates, on the basis of device data including the device state of each production device (101), a value of a plurality of stop factors which affect a decrease in the key performance indicator. The stop factor calculation unit (302): calculates the value of the key performance indicator and the value of the stop factors as values per unit time with respect to values of the same type and of the same time in relation to production on the production line (20); and calculates the value of a non-summed stop factor for which the value of the stop factors has not been calculated on the basis of the device data including the device state of each production device (101). An operator can quantitatively ascertain what is the cause of a decrease in the key performance indicator, and to what degree this cause affects a decrease in a production KPI.

Description

Technical Field

[0001] The present disclosure relates to a production system for analyzing factors causing a production line to stop, a production line analysis method, a learning device, an inference device, a learned model, and a method for generating a learned model.

Background Art

[0002] In a production line where production equipment is arranged, in order to improve the productivity of the production line, it is widely practiced to monitor the operating status of the production line. In improving productivity, for example, there is a method of identifying factors that lower the key performance indicator (KPI), which is an operating index related to defined production, and making improvements. Hereinafter, the key performance indicator related to production is referred to as production KPI.

[0003] Patent Document 1 discloses an operating status monitoring device that selects problem events in the time period around the time when the production KPI has decreased, displays the selected problem events on the operation status transition screen of the display, and performs operations on the display content.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the operating state monitoring device described in Patent Document 1 above, the relationship between the production KPI and the stop factors is only shown on a screen with the time axis as the comparison reference. For this reason, in the operating state monitoring device described in Patent Document 1, there is a problem that in a production line where a plurality of problem events occur simultaneously and in parallel, it is impossible to quantitatively determine how much each stop factor has reduced the production KPI. This problem has become a factor deteriorating the efficiency of the improvement activities to improve the production KPI.

[0006] The present disclosure has been made in view of the above, and an object thereof is to obtain a production system capable of quantifying and presenting a plurality of stop factors that reduce the production KPI of a production line.

Means for Solving the Problems

[0007] In order to solve the above-described problems and achieve the object, a production system according to the present disclosure is a production system that analyzes stop factors of a production line including a plurality of production devices. The production system includes a production result acquisition unit that acquires device data related to the production history of each production device from a plurality of production devices, and from a plurality of production devices, Including identification information for identifying an abnormal state obtained by subdividing the device state of a production device a device state acquisition unit that acquires device data including the device state of each production device, an important performance evaluation index calculation unit that calculates a value of an important performance evaluation index that is a criterion for determining whether the operating status of the production line is good or bad based on the device data related to the production history of each production device, a stop factor calculation unit that calculates values of a plurality of stop factors that affect a decrease in the important performance evaluation index based on the device data including the device state of each production device, and an output unit that displays the value of the important performance evaluation index and the value of the stop factor. The stop factor calculation unit calculates the value of the important performance evaluation index and the value of the stop factor as values per unit time for the same value of the same type of time related to production in the production line, and calculates a value of an unaggregated stop factor for which the value of the stop factor has not been calculated based on the device data including the device state of each production device.

Effects of the Invention

[0008] According to the present disclosure, there is an effect that a plurality of stop factors that reduce the production KPIs of the production line can be quantified and presented.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Embodiments for Carrying Out the Invention

[0010] Hereinafter, the production system, production line analysis method, learning device, inference device, learned model, and method for generating a learned model according to the embodiments will be described in detail with reference to the drawings.

[0011] Embodiment 1. FIG. 1 is a diagram showing the configuration of the overall system 1 having the production system 10 according to Embodiment 1. The overall system 1 includes a production line 20 and a production system 10.

[0012] The production system 10 has a function as a production line analysis system that analyzes line stop factors, which are the stop factors of the production line 20 where a plurality of production devices 101 are connected. The production system 10 collects device data, which is information on the operation of the production device 101, from the production device 101 that constitutes the production line 20, and analyzes the line stop factors, which are the stop factors of the production line 20, based on the device data. The line stop factors correspond to the factors for which the production device 101 has stopped and are factors that affect the decrease in the production KPI of the production line 20. In the following description, the line stop factors may be described as the stop factors of the production device 101, stop factors, or simply factors. The production system 10 includes a data acquisition unit 200 and a data calculation unit 300.

[0013] The production line 20 is an example of a production line that sequentially processes a workpiece, which is a processing object not shown, by a plurality of connected production devices 101. The plurality of production devices 101 in the production line 20 are connected. For this reason, the production capacity of the production line 20 decreases if even one of the production devices 101 stops.

[0014] In FIG. 1, the case where the production apparatuses 101 are arranged one by one in a line is shown. However, the production apparatuses 101 may be arranged such that the arrangement of the production apparatuses 101 branches halfway. For example, the production line 20 may include a configuration in which three (N is a natural number) (N + 1)-th production apparatuses 101 are arranged after the N-th production apparatus 101, and the N-th production apparatus 101 and each of the (N + 1)-th production apparatuses 101 are directly connected to each other. Note that the connection between the production apparatuses 101 in the first embodiment is not limited to the case where the production apparatuses 101 are physically connected by a free-flow conveyor or the like, and also includes the case where the transfer of the work between the production apparatuses 101 is systematically connected by an autonomous robot, an AGV (Automatic Guided Vehicle), a human hand, or the like.

[0015] The production apparatus 101 is various production apparatuses that perform processing on a work. For example, the production apparatus 101 includes, as an example, a soldering apparatus, a coating apparatus, a case assembly apparatus, a screw tightening apparatus, an image inspection apparatus, and the like.

[0016] Each production apparatus 101 includes a production apparatus control unit 1011 that controls the production apparatus 101. The production apparatus control unit 1011 manages time information that is information on time. For example, the production apparatus control unit 1011 holds time data indicating the work processing start time and the work processing end time in the production apparatus 101 as time information. The work processing start time included in the time data is information indicating the date and time when the work processing in the production apparatus 101 started, that is, information indicating the date and time when the production apparatus 101 started the work processing. The work processing end time included in the time data is information indicating the date and time when the work processing in the production apparatus 101 ended, that is, information indicating the date and time when the work processing in the production apparatus 101 was completed. Note that the time from the work processing end time of the first work to the work processing start time of the second work is the waiting time for starting the processing for the second work.

[0017] In addition, the production device 101 individually holds various types of information such as the on or off state of sensors, sensor measurement values, the positions of drive shafts, motor torques, captured images, the display colors of signal towers, the reading results of 2D (Dimensions) code readers, device states, and detailed numbers of device states as device data. An example of the reading result of the 2D code reader is a work ID (Identification) for identifying a work. The device data in Embodiment 1 is associated with time information.

[0018] The network 102 is a communication network for transmitting various device data of the production device 101 to the data acquisition unit 200 located above the production line 20 in the overall system 1. In FIG. 1, the network 102 is shown as a solid line for ease of understanding, but the network 102 may be a wired connection or a wireless connection.

[0019] The data acquisition unit 200 is located above the production line 20 in the overall system 1 and has a function of acquiring various device data of the production device 101. The data acquisition unit 200 is composed of, for example, a computing device such as a personal computer or a PLC (Programmable Logic Controller) and a storage medium such as a database. The data acquisition unit 200 has a production performance acquisition unit 201 and a device state acquisition unit 202, which are functional units realized by the computing device. In addition, the data acquisition unit 200 has a communication unit (not shown) for communicating with the production device 101 and the data calculation unit 300.

[0020] The production result acquisition unit 201 acquires and stores device data related to the production history of the production device 101 from the production device 101. The device data related to the production history of the production device 101 is data on the processing history of the workpiece, which is acquired when the workpiece is processed by the production device 101. That is, the production result acquisition unit 201 acquires various device data such as data on the workpiece processing start time, data on the workpiece processing end time, and data on the pass / fail determination of workpiece processing as the device data related to the production history of the production device 101 from the production device 101. The device data related to the production history of the production device 101 acquired from the production device 101 is used for calculating the production KPI of the production line 20.

[0021] The production KPI is an operation index related to production and is an evaluation index serving as a criterion for judging the quality of the operation status of the production line 20. The higher the value of the production KPI, the better the operation status of the production line 20.

[0022] The device status acquisition unit 202 acquires and stores device data related to the device history of the production device 101 from the production device 101. The device data related to the device history of the production device 101 is data on the history of the state of the production device, which is acquired when the workpiece is processed in the production device 101. That is, the device status acquisition unit 202 acquires various device data such as the display color of the signal tower, the operation mode, and other sensor information as the device data related to the device history of the production device 101 from the production device 101. The device data related to the device history of the production device 101 acquired from the production device 101 is used for calculating the line stop factor, which is the cause of the production line 20 stop, or for analyzing the cause of the line stop factor.

[0023] The data calculation unit 300 is located above the data acquisition unit 200 in the overall system 1 and has a function of quantifying production KPIs and a function of quantifying line stop factors. The data calculation unit 300 is composed of, for example, a computing device such as a personal computer or a programmable logic controller and a storage medium such as a database. The data calculation unit 300 includes a production KPI calculation unit 301 and a stop factor calculation unit 302. Further, the data calculation unit 300 has a communication unit (not shown) that communicates with the data acquisition unit 200 and the output unit 400.

[0024] The production KPI calculation unit 301 has a function of calculating production KPIs using various device data acquired and stored by the data acquisition unit 200 to quantify production KPIs. Specifically, the production KPI calculation unit 301 calculates the value of production KPIs using the device data related to the production history of the production device 101 acquired by the production result acquisition unit 201 of the data acquisition unit 200, and quantifies the production KPIs.

[0025] Note that the production KPI is calculated as a percentage with 100% being the ideal state, and both the numerator and denominator in the calculation formula are converted to time. For example, when the production KPI is the overall equipment efficiency, the definition formula of the production KPI is "number of good products × line tact / operation time". By using this definition formula, the operating status of the production line 20 can be easily grasped.

[0026] The overall equipment efficiency quantifies how much the equipment is actually operating relative to the pre-determined design efficiency within the scope of the equipment's operating schedule.

[0027] The line tact is the time required to produce one product of the same variety. Therefore, the unit of "number of good products × line tact" in the above "number of good products × line tact / operation time" is time. Also, the target line tact, which is the targeted line tact in the production line 20, is used as the line tact.

[0028] The number of non-defective products is the value obtained by counting, without duplicating the work ID, the information indicating that the processing of the work was successful in the pass / fail judgment result of the equipment data related to the production history of the production equipment 101. For example, it is the value obtained by counting, without duplicating the work ID, the information of "〇" in the pass / fail judgment result in FIG. 3 described later.

[0029] Also, when the production KPI is the operation rate, the definition formula of the production KPI is "number of processed items × line tact time / equipment load time". By using this definition formula, the operation status of the production line 20 can be easily grasped. The unit of "number of processed items × line tact time" in "number of processed items × line tact time / equipment load time" is time.

[0030] The equipment load time is the time when the equipment must operate and includes the downtime of the equipment such as the failure time and the setup time.

[0031] The operation rate represents the operation efficiency of the equipment and quantifies the ratio of the time when the equipment operates normally when it is desired to operate the equipment. The operation rate should always aim for 100%. Among "number of processed items × line tact time / equipment load time", "number of processed items × line tact time" corresponds to "the time when the equipment operates normally".

[0032] The stop cause calculation unit 302 has a function of calculating the value of the line stop cause using various equipment data acquired and stored by the data acquisition unit 200 and quantifying the line stop cause. Specifically, the stop cause calculation unit 302 calculates the value of the line stop cause using the equipment data related to the equipment history of the production equipment 101 acquired by the equipment state acquisition unit 202 of the data acquisition unit 200, quantifies the line stop cause, and enables the cause analysis of the line stop cause.

[0033] The line stop cause corresponds to the cause for which the production equipment 101 stopped as described above and is a factor that affects the decrease in the production KPI of the production line 20. The lower the value of the line stop cause, the better the operation status of the production line 20. The value of the line stop cause is calculated as a ratio with the denominator being the same as the production KPI.

[0034] When the production KPI is the overall equipment efficiency, the value of the stop factor is defined as "duration / operating time". In this case, the denominator when calculating the value of the stop factor is the same as the denominator in the calculation formula "number of good products × line tact time / operating time" for calculating the value of the production KPI when the above production KPI is the overall equipment efficiency.

[0035] When the production KPI is the operating rate, the value of the stop factor is defined as "duration / equipment load time". In this case, the denominator when calculating the value of the stop factor is the same as the denominator in the calculation formula "number of processed products × line tact time / equipment load time" for calculating the value of the production KPI when the above production KPI is the operating rate.

[0036] The duration is the time during which the stop state of the production apparatus 101 continues.

[0037] The operating time is the difference between the maximum processing end time and the start time of the production site to which the production line 20 belongs. The maximum processing end time is the latest time among the processing end times of each of the plurality of production apparatuses 101 in the production line 20. For example, if the start time is 8:30, the processing end time of equipment A which is the production apparatus 101 is 17:15, the processing end time of equipment B which is the production apparatus 101 is 17:20, and the processing end time of equipment C which is the production apparatus 101 is 17:30, then 17:30 - 8:30 = 9 hours is defined as the operating time.

[0038] The stop factor calculation unit 302 calculates the value of the unaggregated stop factor for the value of the production KPI calculated as above and the value of the stop factor. The value of the unaggregated stop factor is defined as "100% - value of the production KPI (%) - sum of the values of the plurality of stop factors (%)". That is, the value of the unaggregated stop factor is the remaining value obtained by subtracting the value of the production KPI (%) and the values of the plurality of stop factors (%) from 100%.

[0039] The unaggregated stop factor is a stop factor for which the value of the line stop factor has not been calculated without aggregation in the calculation of the value of the line stop factor using the device data related to the device history of the production device 101 acquired by the device status acquisition unit 202 of the data acquisition unit 200.

[0040] The data calculation unit 300 is realized by, for example, a personal computer. That is, the functions of the production KPI calculation unit 301 and the stop factor calculation unit 302 are realized by, for example, a personal computer introduced with calculation software or a Business Intelligence (BI) tool.

[0041] As described above, in the production system 10, the value of the production KPI, the value of the stop factor, and the value of the unaggregated stop factor are calculated as ratios using the same value in the denominator. That is, in the production system 10, the value of the production KPI, the value of the stop factor, and the value of the unaggregated stop factor are calculated as the value per unit time with respect to the same value of the same type of time related to production on the production line 20. Thereby, in the production system 10, the magnitude relationship between the value of the stop factor and the value of the unaggregated stop factor, and the degree of influence of the stop factor and the unaggregated stop factor on the decrease in the production KPI can be easily compared. Thereby, the operator or the manager can quantitatively grasp what the cause of the decrease in the production KPI is and the degree of influence of the cause on the decrease in the production KPI.

[0042] The output unit 400 is a display unit that can visualize information and present it to the operator at the production site or the manager of the production management department, and is a display device capable of visualizing information, such as a monitor, a tablet, or a wearable device. The output unit 400 can display the changes or numerical values of the quantified production KPI, stop factor, and unaggregated stop factor in a graph or a table. Thereby, the operator or the manager can quantitatively grasp the operating status of the production line 20.

[0043] Next, the operation status analysis process in which the production system 10 quantitatively analyzes and visualizes the operation status of the production line 20 will be described. FIG. 2 is a flowchart showing the processing procedure of the operation status analysis process by the production system 10 according to the first embodiment. Here, the production KPI will be described as the operation rate.

[0044] In step S110, the data acquisition unit 200 acquires and stores the device data of the production line 20 during the processing of the workpiece by the production device 101. That is, the production result acquisition step is performed in which the production result acquisition unit 201 acquires and stores the device data related to the production history of the production device 101 from the production device 101. Further, the device status acquisition unit 202 performs a device status acquisition step of acquiring and storing the device data related to the device history of the production device 101 from the production device 101 while the production device 101 is processing the workpiece.

[0045] FIG. 3 is a diagram showing an example of a first table in which device data used for calculating the operation rate, which is a production KPI, is stored in the production system 10 according to the first embodiment. The device data shown in FIG. 3 is an example of the device data acquired by the production result acquisition unit 201, and is an example of the device data required for the production KPI calculation unit 301 to calculate the operation rate, which is the production KPI. The first table is created by the production result acquisition unit 201 acquiring the device data related to the production history of the production device 101 and storing it in a predetermined format, and is stored in the production result acquisition unit 201.

[0046] In FIG. 3, the "device ID" is device identification information for identifying the production device 101 on the production line 20, and is identification information uniquely assigned to each of the plurality of production devices 101. The "work ID" is work identification information for identifying the work processed by the production device 101, and is identification information uniquely assigned to each of the plurality of works. The "processing start time" is the time when the work starts to be processed by the production device 101. The "processing end time" is the time when the work ends being processed by the production device 101. The "processing time of the work" is calculated from the processing start time and the processing end time of the work. The "pass / fail determination result" is pass / fail information indicating whether the processing of the work by the production device 101 was successful or failed.

[0047] FIG. 4 is a diagram showing an example of a second table in which device data used for calculating the stop factor in the production system 10 according to the first embodiment is stored. The device data shown in FIG. 4 is an example of the device data acquired by the device state acquisition unit 202 and is an example of the device data required for the stop factor calculation unit 302 to calculate the stop factor.

[0048] In FIG. 4, the "device ID" is, as in the case of FIG. 3, device identification information for identifying the production device 101 on the production line 20, and is identification information uniquely assigned to each of the plurality of production devices 101. The "device state" is state information indicating what state the production device 101 was in. The "detail number" is information indicating the state of the production device 101 obtained by further subdividing the device state of the production device 101, and is detailed information obtained by subdividing the "device state". For example, when the "device state" is "abnormal" and the detail number is "46", it means that a trouble corresponding to "error number 46" has occurred in the production device 101. The "state start time" is the time when the state of the production device 101 indicated by the "device state" starts. The "state end time" is the time when the state of the production device 101 indicated by the "device state" ends. The "duration" of the "device state" of the production device 101 is calculated from the state start time and the state end time.

[0049] The error number is identification information for identifying the abnormal state of the production device 101, which is individually assigned to a plurality of abnormal states that can occur in the production device 101.

[0050] The production performance acquisition unit 201 transmits the information of the first table storing the acquired device data to the production KPI calculation unit 301 of the data calculation unit 300. The production KPI calculation unit 301 receives and stores the information of the first table. The device state acquisition unit 202 transmits the information of the second table storing the acquired device data to the stop factor calculation unit 302 of the data calculation unit 300. The stop factor calculation unit 302 receives and stores the information of the second table. Then, it proceeds to step S120.

[0051] In step S120, the data calculation unit 300 calculates the value of the production KPI and the value of the stop factor. Specifically, the production KPI calculation unit 301 of the data calculation unit 300 performs a key performance evaluation index calculation step for calculating the value of the production KPI. Also, the stop factor calculation unit 302 of the data calculation unit 300 performs a stop factor calculation step for calculating the value of the stop factor. The operating rate, which is the production KPI, is defined as "number of processed pieces × line tact time / equipment load time" as described above. And the "line tact time" is a value that can be uniquely determined before the operation of the production line 20. Therefore, the production KPI calculation unit 301 calculates the number of processed workpieces and the equipment load time using the device data as shown in FIG. 3. That is, the production KPI calculation unit 301 acquires the number of processed workpieces and the equipment load time using the device data acquired by the production performance acquisition unit 201 as shown in FIG. 3.

[0052] "Line tact" is input to and stored in the data calculation unit 300 before the operation of the production line 20. The data calculation unit 300 can receive line tact information from outside the data calculation unit 300, for example, via a communication unit. The "number of processed items" is obtained by counting the number of rows in the first table of FIG. 3 without duplication of work IDs. The "equipment load time" can be calculated in the first table of FIG. 3 as the difference between the earliest "processing start time" and the latest "processing end time" in terms of time, targeting the entire information of the first table rather than individual device IDs.

[0053] The value of the stop factor is derived by calculating the "duration for each device state" which is the sum of the differences between the "state start time" and the "state end time" for each row in the second table of FIG. 4, and then dividing the "duration for each device state" by the value of the "equipment load time" used in the calculation of the "operability".

[0054] In this way, the value of the production KPI and the value of the stop factor are calculated by dividing by the same value. That is, the value of the production KPI and the value of the stop factor are calculated as the value per unit time for the same value of the same type of time related to production in the production line 20 by dividing by the value of the same time. Then, the process proceeds to step S130.

[0055] In step S130, the data calculation unit 300 calculates the value of the unaggregated stop factor. Specifically, in step S130, the stop factor calculation unit 302 calculates the value of the unaggregated stop factor. The value of the unaggregated stop factor is calculated by the formula "100% - value of the production KPI (%) - sum of the values of multiple stop factors (%)". Then, the process proceeds to step S140.

[0056] In step S140, the stop factor calculation unit 302 determines whether the value of the unaggregated stop factor calculated in step S130 is a positive value. If the value of the unaggregated stop factor calculated in step S130 is a positive value, then in step S140, it is Yes and the process proceeds to step S150. In this case, the production KPI calculation unit 301 transmits the information on the calculated value of the production KPI to the output unit 400. Also, the stop factor calculation unit 302 transmits the information on the calculated value of the stop factor and the information on the calculated value of the unaggregated stop factor to the output unit 400.

[0057] On the other hand, if the value of the unaggregated stop factor calculated in step S130 is not a positive value, then in step S140, it is No and the process proceeds to step S160. Note that when there is duplication in the data aggregated in FIG. 4, the value of the unaggregated stop factor may become negative. Since the production line 20 is configured by connecting the production apparatuses 101, the stop factor of the adjacent production apparatus 101 may affect the aggregation of the stop factor of the production apparatus 101.

[0058] In step S150, an output step is performed, and the output unit 400 displays the calculated value of the production KPI, the value of the stop factor, and the value of the unaggregated stop factor. Note that when there is no unaggregated stop factor, the output unit 400 does not display the value of the unaggregated stop factor.

[0059] FIG. 5 is a diagram showing a first example of the analysis result of the line stop factor to be displayed by the production system 10 according to Embodiment 1. The graph in FIG. 5 shows the relationship between the line stop factor and the value of the line stop factor in a specific production apparatus 101. In the graph of FIG. 5, as the line stop factor, factor 68, factor 33, unaggregated factor, factor 5, and factor 11 are shown. The output unit 400 displays the value of the stop factor and the value of the unaggregated stop factor as, for example, a bar graph.

[0060] And in the graph of FIG. 5, the values of the stop factors and the values of the unaggregated stop factors are shown as the ratio (%) using the same value in the denominator. As a result, in the graph of FIG. 5, the magnitude relationship between the value of the stop factor and the value of the unaggregated stop factor, and the degree of influence of the stop factor and the unaggregated stop factor on the production KPI can be easily compared. Thereby, the operator or the administrator can quantitatively grasp what the cause of the decrease in the production KPI is and the degree of influence of the cause on the production KPI.

[0061] For example, in FIG. 5, it shows that the stop factor that has the most influence on the decrease in the production KPI is "Factor 68". On the other hand, in FIG. 5, since the unaggregatable stop factor has the third largest value, it suggests that additional investigation or collection of device data is necessary.

[0062] FIG. 6 is a diagram showing a second example of the analysis result of the line stop factors to be displayed by the production system 10 according to Embodiment 1. The graph of FIG. 6 shows the value of the production KPI, the value of the stop factor, and the value of the unaggregated stop factor in one graph, and shows the daily change of the value of the production KPI, the value of the stop factor, and the value of the unaggregated stop factor. That is, FIG. 6 shows the transition of the operating status of the production line 20. In FIG. 6, the horizontal axis indicates the day when the production line 20 stopped, and the vertical axis indicates the value of the production KPI, the value of the stop factor, and the value of the unaggregated stop factor. In the graph of FIG. 6, as the line stop factors, Factor 33, Factor 5, unaggregated factor, Factor 17, and Factor 24 are shown. The output unit 400 displays the change in the value of the production KPI, for example, as a line graph. Further, the output unit 400 displays the change in the value of the stop factor and the value of the unaggregated stop factor, for example, as a bar graph.

[0063] And in the graph of FIG. 6, the values of production KPIs, the values of stop factors, and the values of unaggregated stop factors are displayed as ratios (%) using the same value as the denominator. As a result, in the graph of FIG. 6, the magnitude relationship among the values of production KPIs, the values of stop factors, and the values of unaggregated stop factors, and the degree of influence of stop factors and unaggregated stop factors on production KPIs can be easily compared. Thereby, an operator or a manager can quantitatively grasp what the cause is for the production KPI to decrease and the degree of influence of the cause on the production KPI.

[0064] In step S160, the stop factor calculation unit 302 determines that the unaggregated stop factor is "none" and proceeds to step S150. In this case, the production KPI calculation unit 301 transmits the information on the calculated value of the production KPI to the output unit 400. Further, the stop factor calculation unit 302 transmits the information on the calculated value of the stop factor to the output unit 400. Further, the stop factor calculation unit 302 transmits the information indicating that there is no unaggregated stop factor to the output unit 400.

[0065] Note that the data acquisition unit 200 may be provided in the production apparatus 101. Further, the data acquisition unit 200 and the data calculation unit 300 may be provided in the production apparatus 101.

[0066] According to the production system 10 as described above, a production system for analyzing the causes of stoppage of a production line including a plurality of production devices is realized. The production system includes: a production performance acquisition unit that acquires device data related to the production history of each production device from the plurality of production devices; a device state acquisition unit that acquires device data including the device state of each production device from the plurality of production devices; a key performance indicator calculation unit that calculates the value of a key performance indicator, which is a criterion for determining whether the operating status of the production line is good or bad, based on the device data related to the production history of each production device; a stoppage cause calculation unit that calculates the values of a plurality of stoppage causes that affect the decrease in the key performance indicator, based on the device data including the device state of each production device; and an output unit that displays the value of the key performance indicator and the values of the stoppage causes. The stoppage cause calculation unit realizes a production system that calculates the value of the key performance indicator and the values of the stoppage causes as values per unit time with respect to the same value of the same type of time related to production in the production line.

[0067] As described above, in the production system 10 according to the first embodiment, the production KPI is calculated by the production KPI calculation unit 301 based on the device data related to the production history of the production device 101 acquired by the production performance acquisition unit 201. The value of the production KPI is calculated as a percentage with 100% being the ideal state, and both the numerator and denominator of the calculation formula for calculating the value of the production KPI are converted into time.

[0068] Also, in the production system 10, the value of each stoppage cause is calculated by the stoppage cause calculation unit 302 based on the device data related to the device history of the production device 101 acquired by the device state acquisition unit 202. The value of each stoppage cause is calculated by calculating the duration at each stoppage cause and dividing the duration by the time of the denominator used in the calculation of the production KPI.

[0069] Also, the value of the unaggregated stoppage cause is calculated by "100% - value of the production KPI (%) - total value of the values of each stoppage cause (%)". The value of the production KPI, the values of each stoppage cause, and the value of the unaggregated stoppage cause are displayed on the output unit 400 and presented to the operator or the administrator.

[0070] As described above, in the production system 10, since the values of the production KPI and the stop factors are calculated as ratios using the same value in the denominator, it is possible to quantitatively visualize how much each of the plurality of stop factors has reduced the production KPI. Further, in the production system 10, since it is possible to quantify how much each of the plurality of stop factors has reduced the production KPI, the value of an unaggregated stop factor that is a stop factor reducing the production KPI and has not been aggregated is also calculated as a ratio using the same value as the value of the production KPI and the value of the stop factor in the denominator. Therefore, it is possible to quantitatively visualize how many unaggregated stop factors there are.

[0071] Accordingly, in the production system 10, it is possible to easily compare how much each of the plurality of stop factors has affected the reduction of the production KPI. That is, in the production system 10, it is possible to easily compare the degrees of influence of the plurality of stop factors on the reduction of the production KPI. Then, an operator or a manager at the production site can address each stop factor in an efficient and accurate order to improve the production KPI, and can enhance the efficiency of the improvement activities of the production line 20.

[0072] Therefore, according to the production system 10 according to the first embodiment, there is an effect that a plurality of stop factors that reduce the production KPI of the production line 20 can be quantified and presented.

[0073] Second Embodiment. In the second embodiment, other operation status analysis processing by the production system 10 according to the first embodiment described above will be described. FIG. 7 is a flowchart showing the processing procedure of other operation status analysis processing by the production system 10 according to the first embodiment. Hereinafter, parts different from the flowchart of FIG. 2 described above will be described.

[0074] In other operation status analysis processing, based on the data acquired by the production result acquisition unit 201, after correcting the data acquired by the device status acquisition unit 202, the same processing as the flowchart of FIG. 2 is performed.

[0075] In other operating status analysis processes, for example, using the "device ID" described in FIGS. 3 and 4 as a search key, device states in FIG. 4 that do not fall between the "processing start time" and "processing end time" in FIG. 3 are excluded. Since the production line 20 is configured by connecting the production devices 101, the stop causes of adjacent devices may affect the aggregation of the relevant device. To eliminate such an impact, in other operating status analysis processes, the stop causes are counted only during production processing.

[0076] For example, in the second table shown in FIG. 4, the "device status" of the production device 101 with the "device ID" being "JROC" is "stopped". In this case, the data with the "device status" being "stopped" in the device data of the production device 101 is considered to be the case where the production processing in the production device 101 has stopped due to the stop causes of production devices 101 other than the production device 101. In this case, the data corresponding to the "device status" being "stopped" is excluded and the value of the stop cause is calculated.

[0077] In step S210, the production KPI calculation unit 301 of the data calculation unit 300 calculates the value of the production KPI.

[0078] In step S220, the stop cause calculation unit 302 of the data calculation unit 300 counts the stop causes only for the data during production processing, and calculates the value of the stop cause after limiting to the data during production processing. That is, the stop cause calculation unit 302 acquires the information of the "processing start time" and "processing end time" corresponding to the "device ID" in the first table of FIG. 3, and determines the time from the "processing start time" to the "processing end time" as the "time during production processing" of the "device ID".

[0079] Next, the stop factor calculation unit 302 calculates the value of the stop factor using the second table in FIG. 4. At this time, the stop factor calculation unit 302 searches the second table in FIG. 4 using the "device ID" determined for the "time during production processing" as a search key, and only uses the device data in which the time from the "state start time" to the "state end time" in the second table in FIG. 4 is included in the "time during production processing" determined for the "device ID" to calculate the value of the stop factor. That is, in calculating the value of the stop factor, the stop factor calculation unit 302 excludes the time outside the production processing of the production device 101 from the stop factor time, and totals only the stop factors for the stops caused by the production device 101 that occurred during the production processing of the production device 101.

[0080] In the case of other operation status analysis processes in the above-described Embodiment 2, double counting of stop factors in a plurality of production devices 101 can be prevented. Therefore, in the case of other operation status analysis processes in Embodiment 2, the value of the stop factor can be calculated more accurately than in the case of the operation status analysis process in Embodiment 1 described above. As a result, the operator or administrator can refer to the value of the production KPI calculated by other operation status analysis processes, the value of the stop factor, and the value of the unaggregated stop factor, and take measures against the stop factor in a more accurate order, thereby improving the efficiency of the improvement activities of the production line 20.

[0081] Embodiment 3. Next, Embodiment 3 will be described with reference to FIGS. 8 to 11. In Embodiment 3, when the "production output", for which it is difficult to express the production KPI in terms of time, a machine learning function suitable for cases where various indicators cannot be expressed by a simple mathematical formula, such as when it is difficult to perform the duplicate exclusion calculation of stop factors as described in Embodiment 2, will be described.

[0082] <Learning Phase> FIG. 8 is a diagram showing the configuration of a learning device 50 according to Embodiment 3. The learning device 50 is a computer that learns the degree of influence of a plurality of stop factors in the input state on the decrease in the production KPI.

[0083] The learning device 50 includes a data acquisition unit 51 and a model generation unit 52.

[0084] The data acquisition unit 51 acquires action data and state data as learning data. The data acquisition unit 51 acquires action data and state data from the production system 10 that analyzes the causes of stoppage of the production line 20 including a plurality of production devices 101.

[0085] The action data acquired by the data acquisition unit 51 is action data regarding the degree of influence on the decrease in the production KPI of a plurality of stoppage factors. The state data acquired by the data acquisition unit 51 is state data such as the value of the production KPI, data on the device state of each production device 101, and data on the processing status of each workpiece. The data on the device state of each production device 101 is device data related to the device history of the production device 101 described above, and is device data related to the state of each production device 101 as shown in FIG. 4. The data on the processing status of each workpiece is device data related to the production history of the production device 101 described above, and is device data related to the processing history of each workpiece as shown in FIG. 3. Note that the data on the device state of each production device 101 is not limited to the device data shown in FIG. 4. Also, the data on the processing status of each workpiece is not limited to the device data shown in FIG. 3.

[0086] The model generation unit 52 learns the degree of influence on the decrease in the production KPI of a plurality of stoppage factors based on the learning data including the degree of influence on the decrease in the production KPI of a plurality of stoppage factors, the production KPI, the device state of each production device 101, and the processing status of each workpiece. That is, the model generation unit 52 generates a learned model that infers the degree of influence on the decrease in the production KPI of a plurality of stoppage factors when these are input from the data on the production KPI, the device state of each production device 101, and the processing status of each workpiece in the production system 10. The degree of influence on the decrease in the production KPI of a plurality of stoppage factors in the input state is the degree of influence on the decrease in the production KPI of a plurality of stoppage factors corresponding to the input state data. Hereinafter, the degree of influence on the decrease in the production KPI of a plurality of stoppage factors may be referred to as the degree of influence on the decrease in the production KPI.

[0087] As the learning algorithm used by the model generation unit 52, known algorithms such as supervised learning, unsupervised learning, and reinforcement learning can be used. As an example, the case where the model generation unit 52 applies reinforcement learning to the learning algorithm will be described. In reinforcement learning, an agent (acting entity) in a certain environment observes the current state (parameters of the environment) and determines the action to be taken. The environment dynamically changes due to the agent's actions, and the agent is given a reward according to the change in the environment. The agent repeats this and learns an action policy that can obtain the most rewards through a series of actions. As typical methods of reinforcement learning, Q-learning and TD-learning are known. For example, in the case of Q-learning, the general update formula for the action value function Q(s,a) is represented by the following formula (1).

[0088] [Number]

[0089] In formula (1), s t represents the state of the environment at time t, and a t represents the action at time t. Due to the action a t , the state changes to s t +1. r t+1 represents the reward obtained due to the change in that 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. The degree of influence on the decrease in the production KPI becomes the action a t , the production KPI, the device state of each production device 101, and the processing status of each workpiece become the state s t , and the best action a t in the state s at time t t is learned.

[0090] The update formula represented by Equation (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 represented by Equation (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 sequentially propagated to the action values in the previous environment.

[0091] As described above, when generating a learned model by reinforcement learning, the model generation unit 52 includes a reward calculation unit 53 and a function update unit 54.

[0092] The reward calculation unit 53 calculates a reward based on the degree of influence on the decrease in the production KPI, the production KPI, the device state of each production device 101, and the processing status of each workpiece. The reward calculation unit 53 calculates the reward r based on the difference between the theoretical value of the production KPI and the true value of the production KPI. For example, when the difference between the theoretical value of the production KPI and the true value of the production KPI decreases, the reward r is increased (for example, a reward of "1" is given), while when the difference between the theoretical value of the production KPI and the true value of the production KPI increases, the reward r is decreased (for example, a reward of "-1" is given).

[0093] The theoretical value of the production KPI is the value of the production KPI calculated by 100% - Σ (degree of influence on the decrease in the production KPI × duration of the stop factor). Σ (degree of influence on the decrease in the production KPI × duration of the stop factor) is the total value of the values of (degree of influence on the decrease in the production KPI × duration of the stop factor) for each stop factor.

[0094] The true value of the production KPI is, for example, the value of the production KPI calculated from the device data shown in Figure 3 itself. Note that as the definition of the production KPI, any appropriate value may be selected according to the operation mode of each production site.

[0095] The function update unit 54 updates a function for determining the degree of influence on the decrease in the production KPI in the input state according to the reward calculated by the reward calculation unit 53, and outputs it to the learned model storage unit 70 as a learned model. For example, in the case of Q-learning, the function update unit 54 uses the action value function Q(s t ,a t ) as a function for calculating the degree of influence on the decrease in the production KPI in the input state.

[0096] The model generation unit 52 repeatedly executes the above learning. The learned model storage unit 70 stores the action value function Q(s t ,a t ) updated by the function update unit 54, that is, stores the learned model.

[0097] Next, with reference to FIG. 9, the process by which the learning device 50 learns the degree of influence on the decrease in the production KPI will be described. FIG. 9 is a flowchart showing the processing procedure of the learning process by the learning device 50 according to Embodiment 3.

[0098] In step S310, the data acquisition unit 51 acquires, as learning data, the degree of influence on the decrease in the production KPI, the production KPI, the device state of each production device 101, and the processing status of each workpiece.

[0099] In step S320, the model generation unit 52 calculates a reward based on the degree of influence on the decrease in the production KPI, the production KPI, the device state of each production device 101, and the processing status of each workpiece. Specifically, the reward calculation unit 53 acquires the degree of influence on the decrease in the production KPI, the production KPI, the device state of each production device 101, and the processing status of each workpiece, and determines whether to increase or decrease the reward based on the difference between the theoretical value of the production KPI determined in advance and the true value of the production KPI.

[0100] When the reward calculation unit 53 determines that the reward should be increased (in step S320, the difference between the theoretical value and the true value of the production KPI decreases), it increases the reward in step S330. That is, when the difference between the theoretical value and the true value of the production KPI decreases and the reward increase criterion is satisfied, the reward calculation unit 53 increases the reward.

[0101] On the other hand, when the reward calculation unit 53 determines that the reward should be decreased (in step S320, the difference between the theoretical value and the true value of the production KPI increases), it decreases the reward in step S340. That is, when the difference between the theoretical value and the true value of the production KPI increases and the reward decrease criterion is satisfied, the reward calculation unit 53 decreases the reward.

[0102] In step S350, the function update unit 54 updates the action value function Q(s t , a t ) represented by formula (1) stored in the learned model storage unit 70 based on the reward calculated by the reward calculation unit 53.

[0103] The learning device 50 repeatedly executes the steps from step S310 to step S350 above, and stores the generated action value function Q(s t , a t ) in the learned model storage unit 70 as a learned model.

[0104] Step S310 can be said to be a data acquisition step. Steps S320 to S350 can be said to be a model generation step. Then, the learning device 50 repeatedly executes the above data acquisition step and model generation step, updates the action value function Q(s t , a t ) represented by formula (1) stored in the learned model storage unit 70, and stores it in the learned model storage unit 70.

[0105] Although the learning device 50 according to Embodiment 3 has been described for the case of storing the learned model in the learned model storage unit 70 provided outside the learning device 50, the learned model storage unit 70 may be arranged inside the learning device 50.

[0106] According to the learning device 50 according to Embodiment 3, from a production system that analyzes the causes of stoppage of a production line including a plurality of production devices, the important performance evaluation index of the production line, the device states of the plurality of production devices, the processing status of the workpiece processed by the production device, and the importance of the stoppage factor in the important performance evaluation index, the device states of the plurality of production devices, and the processing status of the workpiece. A data acquisition unit that acquires learning data including the degree of influence on the reduction of the performance evaluation index, and using the learning data, the important performance evaluation index of the production line, the device states of the plurality of production devices, and the processing status of the workpiece. A learning device including a model generation unit that generates a learned model for inferring the degree of influence on the reduction of the important performance evaluation index of the stoppage factor is realized.

[0107] According to the learning device 50 according to Embodiment 3, from a production system that analyzes the causes of stoppage of a production line including a plurality of production devices, the important performance evaluation index of the production line, the device states of the plurality of production devices, the processing status of the workpiece processed by the production device, and the importance of the stoppage factor in the important performance evaluation index, the device states of the plurality of production devices, and the processing status of the workpiece. A data acquisition step of acquiring learning data including the degree of influence on the reduction of the performance evaluation index, and using the learning data, the important performance evaluation index of the production line, the device states of the plurality of production devices, and the processing status of the workpiece. A method for generating a learned model including a model generation step of generating a learned model for inferring the degree of influence on the reduction of the important performance evaluation index of the stoppage factor is implemented.

[0108] According to the learning device 50 according to Embodiment 3, a method for generating a learned model that updates the learned model is implemented by performing a new data acquisition step of acquiring new learning data and performing a new model generation step using the new learning data.

[0109] According to the learning device 50 according to Embodiment 3, using learning data including the important performance evaluation indicators of the production line, the device states of a plurality of production devices, the processing status of the workpieces processed by the production devices, and the degree of influence on the decrease in the important performance evaluation indicators of the stop factors in the important performance evaluation indicators, the device states of the plurality of production devices, and the processing status of the workpieces, machine learning is performed to generate a learned model that outputs the degree of influence on the decrease in the important performance evaluation indicators of the stop factors from the important performance evaluation indicators, the device states of the plurality of production devices, and the processing status of the workpieces.

[0110] <Utilization Phase> FIG. 10 is a diagram showing the configuration of the inference device 60 according to Embodiment 3. The inference device 60 is a computer that infers the degree of influence on the decrease in the production KPI using a learned model.

[0111] The inference device 60 includes a data acquisition unit 61 and an inference unit 62.

[0112] The data acquisition unit 61 acquires the production KPI, the device states of the respective production devices 101, and the processing status of the respective workpieces from the production system 10 that analyzes the stop factors of the production line 20 including the plurality of production devices 101.

[0113] The inference unit 62 infers the degree of influence on the decrease in the production KPI in the input state using the learned model stored in the learned model storage unit 70. That is, the inference unit 62 can infer the degree of influence on the decrease in the production KPI suitable for the input production KPI, the device states of the respective production devices 101, and the processing status of the respective workpieces by inputting the production KPI, the device states of the respective production devices 101, and the processing status of the respective workpieces acquired by the data acquisition unit 61 into this learned model.

[0114] In addition, in the third embodiment, the inference device 60 is described as outputting the degree of influence on the decrease in production KPIs using the learned model learned by the model generation unit 52 for the production system 10 connected to the production line 20. However, the inference device 60 may acquire a learned model learned for another production system connected to another production line. In this case, the inference device 60 outputs the degree of influence on the decrease in production KPIs using the learned model learned for another production system connected to another production line.

[0115] Next, with reference to FIG. 11, the process by which the inference device 60 infers the degree of influence on the decrease in production KPIs will be described. FIG. 11 is a flowchart showing the processing procedure of the inference process by the inference device 60 according to the third embodiment.

[0116] In step S410, the data acquisition unit 61 acquires, as inference data, data on production KPIs, the device states of the respective production devices 101, and the processing status of each workpiece.

[0117] In step S420, the inference unit 62 inputs the data on production KPIs, the device states of the respective production devices 101, and the processing status of each workpiece into the learned model stored in the learned model storage unit 70, and obtains the degree of influence on the decrease in production KPIs corresponding to the input information. The degree of influence on the decrease in production KPIs obtained here outputs the degree of influence on the decrease in production KPIs for each of a plurality of stop factors individually. For example, for stop factor A, the production KPI is decreased by P%, and for stop factor B, the production KPI is decreased by Q%. Thus, the degree of influence on the decrease in production KPIs for each of a plurality of stop factors is output individually. Also, the unit of the degree of influence on the decrease in production KPIs is not limited to "%". For example, when the production quantity is used as the production KPI, the degree of influence on the decrease in production KPIs may be such that for stop factor A, the production quantity is decreased by R, and for stop factor B, the production quantity is decreased by S, with the number being the unit. In step S430, the inference unit 62 outputs the obtained degree of influence on the decrease in production KPIs to the output unit 400 of the production system 10.

[0118] In step S430, the output unit 400 of the production system 10 outputs the degree of influence on the decrease in the production KPI sent from the inference unit 62, that is, by displaying the degree of influence on the decrease in the production KPI, the operator at the production site is notified of the degree of influence on the decrease in the production KPI.

[0119] In step S440, each production device 101 performs the processing of the work under the operating conditions of the production device 101 that improves the production KPI of the production line 20. That is, the operator at the production site checks the degree of influence on the decrease in the production KPI displayed on the output unit 400, conducts improvement activities to improve the production KPI of the production line 20, determines the improved operating conditions that are the operating conditions of the production device 101 for improving the production KPI of the production line 20, and sets the determined improved operating conditions in each production device 101. Each production device 101 performs the processing of the work under the newly set improved operating conditions of the production device 101 that improves the production KPI of the production line 20.

[0120] According to the inference device 60 according to the third embodiment, from a production system that analyzes the stop factors of a production line including a plurality of production devices, a data acquisition unit that acquires the important performance evaluation index of the production line, the device states of the plurality of production devices, and the processing status of the work processed by the production devices, and a learned model for inferring the degree of influence on the decrease in the important performance evaluation index of the stop factor in the important performance evaluation index of the production line, the device states of the plurality of production devices, and the processing status of the work, the inference unit outputs the degree of influence on the decrease in the important performance evaluation index of the stop factor from the important performance evaluation index, the device states of the plurality of production devices, and the processing status of the work acquired by the data acquisition unit. An inference device including the above is realized.

[0121] By applying the above-described learning device 50 and inference device 60 to the production system 10, the inference device 60 can automatically present the degree of influence on the decrease in the production KPI of the stop factor to the operator or manager at the production site corresponding to the stop status of each production device 101. Then, the operator at the production site refers to the degree of influence on the decrease in the production KPI of the stop factor of the production device 101 and proceeds with the improvement activities, thereby enhancing the efficiency of the production line improvement activities.

[0122] In addition, in the third embodiment, although the case where reinforcement learning is applied to the learning algorithm used by the inference unit 62 has been described, the learning algorithm is not limited to reinforcement learning. Regarding the learning algorithm used by the inference unit 62, in addition to reinforcement learning, supervised learning, unsupervised learning, semi-supervised learning, or the like can also be applied.

[0123] Also, as the learning algorithm used for the model generation unit 52, deep learning that learns to extract the feature quantity itself can be used. Further, the model generation unit 52 may execute machine learning according to other known methods, such as neural networks, genetic programming, functional logic programming, support vector machines, and the like.

[0124] Note that the learning device 50 and the inference device 60 may be devices separate from the production system 10 connected to the production system 10 via a network such as the Internet, for example. Further, the learning device 50 and the inference device 60 may be built into the production system 10. Furthermore, the learning device 50 and the inference device 60 may exist on a cloud server.

[0125] Further, the model generation unit 52 may learn the degree of influence on the decrease in the production KPI in the input state using the learning data acquired from the plurality of production systems 10. Note that the model generation unit 52 may acquire learning data from a plurality of production systems 10 used in the same area, or may learn the degree of influence on the decrease in the production KPI in the input state using the learning data collected from a plurality of production systems 10 operating independently in different areas. Also, the production system 10 that collects the learning data may be added to the target midway or removed from the target. Further, the learning device 50 that has learned the degree of influence on the decrease in the production KPI in the state input to a certain production system 10 may be applied to another production system 10 different from this, and the degree of influence on the decrease in the production KPI in the state input to the other production system 10 may be relearned and updated.

[0126] Here, the hardware configurations of the learning device 50 and the inference device 60 will be described. Since the learning device 50 and the inference device 60 have the same hardware configuration, the hardware configuration of the learning device 50 will be described here.

[0127] The learning device 50 is realized by a processing circuit. The processing circuit may be a processor and a memory that execute a program stored in the memory, or may be dedicated hardware such as a dedicated circuit. The processing circuit is also called a control circuit.

[0128] FIG. 12 is a diagram showing a configuration example of a processing circuit included in the learning device 50 according to Embodiment 3 when the processing circuit is realized by a processor and a memory. The processing circuit 90 shown in FIG. 12 is a control circuit and includes a processor 91 and a memory 92. When the processing circuit 90 is composed of 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, the processor 91 reads and executes the program stored in the memory 92 to realize each function. That is, the processing circuit 90 includes a memory 92 for storing a program that will ultimately execute the processing of the learning device 50. 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 a learning process.

[0129] Here, the processor 91 is, for example, a CPU (Central Processing Unit), a processing device, an arithmetic device, a microprocessor, a microcomputer, or a DSP (Digital Signal Processor). The processor 91 is included in a PC (Personal Computer) or a PLC. The PLC is also called a sequencer.

[0130] In addition, the memory 92 corresponds to, for example, non-volatile or volatile semiconductor memories such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable ROM), EEPROM (registered trademark) (Electrically EPROM), magnetic disks, flexible disks, optical disks, compact disks, mini disks, or DVDs (Digital Versatile Discs).

[0131] FIG. 13 is a diagram showing an example of a processing circuit when the processing circuit included in the learning device 50 according to the third embodiment is configured by dedicated hardware. The processing circuit 93 shown in FIG. 13 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 thereof. For the processing circuit 93, a part may be realized by dedicated hardware and a part may be realized by software or firmware. Thus, the processing circuit 93 can realize each of the above functions by dedicated hardware, software, firmware, or a combination thereof.

[0132] Note that the production result acquisition unit 201, the device state acquisition unit 202, the production KPI calculation unit 301, and the stop factor calculation unit 302 also have the hardware configurations described with reference to FIGS. 12 and 13.

[0133] Embodiment 4. In Embodiment 4, a case where the overall system 1 according to Embodiment 1 described above includes the corresponding operator proposal unit 500 will be described. FIG. 14 is a diagram showing the configuration of the overall system 1 having the corresponding operator proposal unit 500 according to Embodiment 4.

[0134] The corresponding operator proposal unit 500 is located below the data calculation unit 300 in the overall system 1. When the change rate of the value of the line stop factor calculated by the stop factor calculation unit 302 of the data calculation unit 300 exceeds a preset threshold value, the line stop factor is determined as the stop factor to be removed, and it has a function of automatically determining the corresponding operator who is the operator corresponding to the removal of the stop factor to be removed among the plurality of operators at the production site.

[0135] The corresponding operator proposal unit 500 includes a current stop factor grasping unit 501, an operator position information acquisition unit 502, and a corresponding operator determination unit 503. The corresponding operator proposal unit 500 is composed of, for example, a computing device such as a personal computer or a programmable logic controller, and a storage medium such as a database. In addition, the corresponding operator proposal unit 500 has information acquisition devices such as an overhead camera and a beacon receiver. Further, the corresponding operator proposal unit 500 has a communication unit (not shown) that communicates with the data acquisition unit 200 and the output unit 400.

[0136] The current stop factor grasping unit 501 detects at any time the change in the influence degree of the line stop factor on the decrease of the important performance evaluation index. The current stop factor grasping unit 501 acquires the value of the line stop factor from the stop factor calculation unit 302 of the data calculation unit 300 for a plurality of production devices 101, and determines the line stop factor whose change rate of the value of the line stop factor exceeds a preset threshold value as the stop factor to be removed. The threshold value is a threshold value for the current stop factor grasping unit 501 to determine the stop factor to be removed by comparing with the change rate of the value of the line stop factor.

[0137] That is, the current stop factor grasping unit 501 has a function of acquiring the value (%) of each line stop factor calculated by the stop factor calculation unit 302 of the data calculation unit 300 from the stop factor calculation unit 302 at a predetermined cycle for a plurality of production devices 101 on the production line 20, and monitoring the change rate of the acquired value of the line stop factor. That is, the current stop factor grasping unit 501 has a function of constantly grasping the current line stop factor of each production device 101 at the production site.

[0138] Then, when the change rate of the value of the line stop factor exceeds a preset threshold, the current stop factor grasping unit 501 determines that it is necessary to remove the line stop factor. That is, when the change rate of the value of the line stop factor exceeds a preset threshold, the current stop factor grasping unit 501 determines that the line stop factor is a line stop factor to be removed, which is a line stop factor that needs to be removed.

[0139] The change rate is a value calculated by taking an arbitrary time as the denominator and the variation value (%) of the line stop factor within the arbitrary time as the numerator. When the change rate of a certain line stop factor is relatively large compared to the change rates of other line stop factors, it can be said that the line stop factor is currently occurring and the impact of the line stop factor on the production KPI is relatively large.

[0140] The operator position information acquisition unit 502 determines the positions of a plurality of operators at the production site where a plurality of production apparatuses 101 are installed. That is, the operator position information acquisition unit 502 has a function of determining and storing the positions of each operator at the production site where the production line 20 is installed. The operator position information acquisition unit 502 has devices such as, for example, an overhead camera provided on the ceiling of the production site and a beacon receiver for acquiring information on the positions of beacons carried by the operators. The operator position information acquisition unit 502 determines and stores the positions of each operator at the production site based on, for example, an image of the overhead camera provided on the ceiling of the production site. Also, the operator position information acquisition unit 502 determines and stores the positions of each operator at the production site based on, for example, information received by the beacon receiver.

[0141] The corresponding worker determination unit 503 determines the corresponding worker who is responsible for removing the line stop factor to be removed among the multiple workers present at the production site where multiple production devices 101 are installed. That is, the corresponding worker determination unit 503 has a function of determining and proposing which worker among the workers present at the production site should be directed to remove the line stop factor that is detected by the current stop factor grasping unit 501 and is reducing the production KPI. The corresponding worker determination unit 503 holds the position information of the production device 101 and the stop factor attribute information.

[0142] The position information of the production device 101 is used to calculate the distance between the production device 101 and each worker by combining it with the position information of the worker acquired by the worker position information acquisition unit 502. That is, the corresponding worker determination unit 503 acquires the position information of each worker from the worker position information acquisition unit 502. Then, the corresponding worker determination unit 503 calculates the distance between the production device 101 and each worker based on the position information of each worker and the held position information of the production device 101.

[0143] The stop factor attribute information holds, as data, which production device 101 each line stop factor belongs to. Also, each line stop factor is associated with the hat color photographed by the overhead camera and the worker number set in the beacon. That is, the stop factor attribute information includes the correspondence information between the hat color and the line stop factor indicating which worker with what hat color is in charge of each line stop factor. Also, the stop factor attribute information includes the correspondence information between the worker number set in the beacon and the line stop factor indicating which worker is in charge of each line stop factor.

[0144] Subsequently, the operation of the corresponding worker proposal unit 500 will be described.

[0145] During the operation of production line 20, the current stoppage cause grasping unit 501 acquires the values of the line stoppage causes from the data calculation unit 300 and constantly monitors the change rate of the values of the line stoppage causes. When the change rate of the value of the line stoppage cause exceeds a preset threshold during the monitoring of the change rate of the value of the line stoppage cause by the current stoppage cause grasping unit 501, it is determined that the line stoppage cause needs to be removed. That is, when the change rate of the value of the line stoppage cause exceeds a preset threshold, the current stoppage cause grasping unit 501 determines that the line stoppage cause is a line stoppage cause to be removed.

[0146] The corresponding operator determination unit 503 identifies which production device 101 the line stoppage cause that has been determined to need to be removed, that is, the line stoppage cause to be removed, has occurred in. That is, the corresponding operator determination unit 503 identifies which production device 101 the line stoppage cause to be removed has occurred in based on the held stoppage cause attribute information.

[0147] Also, the corresponding operator determination unit 503 identifies the operator who can handle the removal of the line stoppage cause based on the held stoppage cause attribute information, and calculates the distance between each identified operator and the production device 101. For calculating the distance between the operator and the production device 101, the corresponding operator determination unit 503 acquires and uses the information on the position of the operator at the time of calculation from the operator position information acquisition unit 502. The corresponding operator determination unit 503 determines that the operator with the shortest calculated distance is the corresponding operator who should handle the removal of the line stoppage cause.

[0148] The corresponding operator proposal unit 500 causes the output unit 400 to display the name of the operator who should handle the removal of the line stop factor. For example, when the output unit 400 is composed of a large monitor and a wearable device, the corresponding operator proposal unit 500 displays the ongoing line stop factor and the name of the operator who should handle the removal of the line stop factor side by side. In addition, the corresponding operator proposal unit 500 notifies the operator to instruct the operator to move to the production device 101 where the stop factor has occurred. Examples of the notification method to the operator include a method of vibrating the operator's wearable device in a predetermined pattern and a method of sounding a predetermined sound with the operator's wearable device.

[0149] Note that the line stop factor and the operator name displayed on the large monitor of the output unit 400 can be deleted at any timing, such as when the line stop factor is resolved or when it is notified from the operator's wearable device to the corresponding operator proposal unit 500 that the line stop factor cannot be handled. When the line stop factor continues and only the operator name is deleted from the display on the large monitor, the corresponding operator determination unit 503 re-performs the determination with the operator excluded from the operators who should handle the removal of the line stop factor, and determines an alternative corresponding operator.

[0150] Such an overall system 1 equipped with the corresponding operator proposal unit 500 can automatically and at any time propose the operator who should handle each line stop factor that deteriorates the production KPI, so it is possible to suppress the duration of each line stop factor.

[0151] The configuration shown in the above embodiments is an example, and it is possible to combine it with another known technology, combine the embodiments with each other, and omit or change a part of the configuration without departing from the gist.

[0152] Hereinafter, various aspects of the present disclosure will be collectively described as appendices.

[0153] (Appendix 1) A production line analysis system for analyzing the causes of stoppage of a production line including a plurality of production devices, a production performance acquisition unit that acquires device data related to the production history of each of the plurality of production devices from the plurality of production devices, a device state acquisition unit that acquires device data including the device state of each of the plurality of production devices from the plurality of production devices, a key performance evaluation index calculation unit that calculates a value of a key performance evaluation index, which is a criterion for determining whether the operating status of the production line is good or bad, based on the device data related to the production history of each production device, a stoppage cause calculation unit that calculates values of a plurality of the stoppage causes that affect a decrease in the key performance evaluation index based on the device data including the device state of each production device, an output unit that displays the value of the key performance evaluation index and the value of the stoppage cause, comprising, the stoppage cause calculation unit calculates the value of the key performance evaluation index and the value of the stoppage cause as values per unit time with respect to the same value of the same type of time related to production in the production line, A production line analysis system characterized by the above. (Appendix 2) the stoppage cause calculation unit calculates the value of an unaggregated stoppage cause for which the value of the stoppage cause has not been calculated based on the device data including the device state of each production device by a calculation formula of "100 - key performance evaluation index - total stoppage cause value", The production line analysis system according to Appendix 1, characterized by the above. (Appendix 3) the output unit displays the value of the stoppage cause and the value of the unaggregated stoppage cause on the same graph, The production line analysis system according to Appendix 2, characterized by the above. (Appendix 4) the output unit displays the value of the key performance evaluation index, the value of the stoppage cause, and the value of the unaggregated stoppage cause on the same graph, The production line analysis system according to Appendix 2, characterized by the above. (Appendix 5) The device data related to the production history of each production device is device identification information for identifying the production device, work identification information for identifying the work processed by the production device, the work processing start time by the production device, the work processing end time by the production device, pass / fail information indicating whether the work processing by the production device was successful or failed, and includes The device data including the device state of each production device is device identification information for identifying the production device, status information indicating what state the production device was in, detailed information obtained by subdividing the device state, the state start time which is the time when the state of the production device indicated by the device state started, the state end time which is the time when the state of the production device indicated by the device state ended, and includes The production line analysis system according to any one of Appendices 1 to 4, characterized by the above. (Appendix 6) The definition formula of the important performance evaluation index is line tact × number of processed items / equipment load time, where the line tact is the target tact targeted in the production line, the number of processed items is the value obtained by counting the work identification information in the device data including the device state of each production device without duplication, the equipment load time is the difference between the earliest work processing start time and the latest work processing end time over the entire device data related to the production history of each production device, The production line analysis system according to Appendix 5, characterized by the above. (Appendix 7) The definition formula of the important performance evaluation index is number of non-defective items × line tact / operation time, where the line tact is the target tact targeted in the production line, The number of good products is a value obtained by counting, without duplication, the work identification information for information indicating successful processing of the work in the pass / fail information of the equipment data related to the production history of each production device. The operating time is the difference between the maximum processing end time and the start time of the production site to which the production line belongs. The production line analysis system according to supplementary note 5, characterized by the above. (Supplementary note 8) In calculating the value of the stop factor, the stop factor calculation unit excludes the time outside the production processing of the production device. The production line analysis system according to any one of supplementary notes 1 to 7, characterized by the above. (Supplementary note 9) A production line analysis method for analyzing the stop factors of a production line including a plurality of production devices, A production performance acquisition step of acquiring equipment data related to the production history of each production device from a plurality of the production devices, An equipment state acquisition step of acquiring equipment data including the equipment state of each production device from a plurality of the production devices, An important performance evaluation index calculation step of calculating a value of an important performance evaluation index, which is a criterion for determining the quality of the operating status of the production line, based on the equipment data related to the production history of each production device, A stop factor calculation step of calculating values of a plurality of the stop factors that affect a decrease in the important performance evaluation index based on the equipment data including the equipment state of each production device, An output step of displaying the value of the important performance evaluation index and the value of the stop factor, including In the stop factor calculation step, the value of the important performance evaluation index and the value of the stop factor are calculated as values per unit time for the same value of the same type of time related to production in the production line. The production line analysis method characterized by the above. (Supplementary note 10) From a production line analysis system that analyzes the causes of stoppage of a production line including a plurality of production devices, a key performance evaluation index of the production line, the device states of the plurality of production devices, the processing status of workpieces processed by the production devices, and the degree of influence on the reduction of the key performance evaluation index of the stoppage cause in the key performance evaluation index, the device states of the plurality of production devices, and the processing status of the workpieces, a data acquisition unit that acquires learning data including A model generation unit that generates a learned model for inferring the degree of influence on the reduction of the key performance evaluation index of the stoppage cause from the key performance evaluation index of the production line, the device states of the plurality of production devices, and the processing status of workpieces processed by the production devices using the learning data A learning device, characterized by comprising (Appendix 11) A data acquisition unit that acquires from a production line analysis system that analyzes the causes of stoppage of a production line including a plurality of production devices, the key performance evaluation index of the production line, the device states of the plurality of production devices, and the processing status of workpieces processed by the production devices An inference unit that outputs the degree of influence on the reduction of the key performance evaluation index of the stoppage cause from the key performance evaluation index of the production line, the device states of the plurality of production devices, and the processing status of workpieces processed by the production devices acquired by the data acquisition unit, using a learned model for inferring the degree of influence on the reduction of the key performance evaluation index of the stoppage cause in the key performance evaluation index, the device states of the plurality of production devices, and the processing status of the workpieces An inference device, characterized by comprising

Explanation of reference signs

[0154] 1 Whole system, 10 Production system, 20 Production line, 50 Learning device, 51, 61, 200 Data acquisition unit, 52 Model generation unit, 53 Reward calculation unit, 54 Function update unit, 60 Inference device, 62 Inference unit, 70 Trained model storage unit, 90, 93 Processing circuit, 91 Processor, 92 Memory, 101 Production device, 102 Network, 201 Production performance acquisition unit, 202 Device state acquisition unit, 300 Data calculation unit, 301 Production KPI calculation unit, 302 Stop factor calculation unit, 400 Output unit, 500 Corresponding worker proposal unit, 501 Current stop factor grasping unit, 502 Worker position information acquisition unit, 503 Corresponding worker determination unit, 1011 Production device control unit.

Claims

1. A production system for analyzing the causes of stoppage of a production line including a plurality of production devices, a production performance acquisition unit that acquires device data related to the production history of each of the plurality of production devices from the plurality of production devices, a device state acquisition unit that acquires device data including the device state of each of the plurality of production devices, the device state including identification information for identifying an abnormal state obtained by subdividing the device state of the production device, from the plurality of production devices, a key performance evaluation index calculation unit that calculates a value of a key performance evaluation index, which is a criterion for determining whether the operating status of the production line is good or bad, based on the device data related to the production history of each of the production devices, a stoppage cause calculation unit that calculates values of a plurality of the stoppage causes that affect a decrease in the key performance evaluation index based on the device data including the device state of each of the production devices, an output unit that displays the value of the key performance evaluation index and the value of the stoppage cause, comprising, wherein the stoppage cause calculation unit, calculates the value of the key performance evaluation index and the value of the stoppage cause as values per unit time for the same value of the same type of time related to production in the production line, calculates a value of an unaggregated stoppage cause for which the value of the stoppage cause has not been calculated based on the device data including the device state of each of the production devices, characterized in that it is a production system.

2. The stoppage cause calculation unit calculates the value of the unaggregated stoppage cause by a calculation formula of "100 - key performance evaluation index - total stoppage cause value", characterized in that it is the production system according to claim 1.

3. The output unit displays the value of the stoppage cause and the value of the unaggregated stoppage cause on the same graph, characterized in that it is the production system according to claim 2.

4. The output unit displays the value of the key performance evaluation index, the value of the stoppage cause, and the value of the unaggregated stoppage cause on the same graph, characterized in that it is the production system according to claim 2.

5. The device data related to the production history of each of the production devices includes, device identification information for identifying the production device, work identification information for identifying the work processed by the production device, the work processing start time by the production device, the work processing end time by the production device, pass / fail information indicating whether the work processing by the production device was successful or failed, including, the device data including the device state of each of the production devices includes, device identification information for identifying the production device, Status information indicating the state of the production device, Detailed information obtained by subdividing the device state, including identification information for identifying abnormal states obtained by subdividing the device state of the production device, A state start time, which is the time when the state of the production device indicated by the device state is started, A state end time, which is the time when the state of the production device indicated by the device state is ended, Including, The production system according to claim 1, characterized in that.

6. The definition formula of the important performance evaluation index is line tact × number of processed items / equipment load time, The line tact is the target line tact targeted in the production line, The number of processed items is a value obtained by counting the work identification information in the device data related to the production history of each production device without duplication, The equipment load time is the difference between the earliest processing start time and the latest processing end time in terms of time, targeting the entire device data related to the production history of each production device, The production system according to claim 5, characterized in that.

7. The definition formula of the important performance evaluation index is number of good products × line tact / operating time, The line tact is the target line tact targeted in the production line, The number of good products is a value obtained by counting the information indicating successful processing of the work in the pass / fail information of the device data related to the production history of each production device without duplication of the work identification information, The operating time is the difference between the latest processing end time and the start time of the production site to which the production line belongs, The production system according to claim 5, characterized in that.

8. The stop cause calculation unit excludes the time outside the production processing of the production device in calculating the value of the stop cause, The production system according to claim 1, characterized in that.

9. A current stop cause grasping unit that acquires the value of the stop cause from the stop cause calculation unit for a plurality of the production devices and determines the stop cause whose change rate of the value of the stop cause exceeds a preset threshold as the stop cause to be removed; An operator position information acquisition unit that determines the positions of a plurality of operators at a production site where a plurality of the production devices are installed; A corresponding operator determination unit that determines a corresponding operator who is an operator corresponding to the removal of the stop cause to be removed among the plurality of operators; The production system according to any one of claims 1 to 8, characterized by comprising.

10. Based on the stop factor attribute information indicating to which of the production devices the stop factor belongs and the position information of the production device, calculate the distance between the production device and a plurality of the workers, and determine the worker with the shortest calculated distance as the corresponding worker. The production system according to claim 9, characterized in that.

11. A production line analysis method for analyzing the stop factors of a production line including a plurality of production devices, A production performance acquisition step of acquiring device data related to the production history of each production device from a plurality of the production devices, A device state acquisition step of acquiring device data including the device state of each production device including identification information for identifying an abnormal state obtained by subdividing the device state of the production device from a plurality of the production devices, An important performance evaluation index calculation step of calculating the value of an important performance evaluation index which is a criterion for determining whether the operation status of the production line is good or bad based on the device data related to the production history of each production device, A stop factor calculation step of calculating the values of a plurality of the stop factors that affect the decrease of the important performance evaluation index based on the device data including the device state of each production device, An output step of displaying the value of the important performance evaluation index and the value of the stop factor, including, In the stop factor calculation step, calculate the value of the important performance evaluation index and the value of the stop factor as the value per unit time for the same value of the same type of time related to production in the production line, calculate the value of an unaggregated stop factor for which the value of the stop factor has not been calculated based on the device data including the device state of each production device, A production line analysis method characterized by the above.

12. From a production system for analyzing the stop factors of a production line including a plurality of production devices, an important performance evaluation index which is a criterion for determining whether the operation status of the production line is good or bad, the device states of a plurality of the production devices including identification information for identifying an abnormal state obtained by subdividing the device state of the production device, the processing status of the workpiece processed by the production device, and the degree of influence on the decrease of the important performance evaluation index of the stop factor in the important performance evaluation index, the device states of a plurality of the production devices, and the processing status of the workpiece, a data acquisition unit for acquiring learning data including. A model generation unit that generates a learned model for inferring the degree of influence on the decline in the key performance evaluation index of the stop factor from the key performance evaluation index of the production line, the device states of a plurality of the production devices, and the processing status of the work processed by the production device using the learning data; comprising; The learned model individually outputs the degree of influence on the decline in the key performance evaluation index of the stop factor for a plurality of the stop factors; A learning device characterized by this.

13. From a production system that analyzes the stop factors of a production line including a plurality of production devices, the key performance evaluation index that is the criterion for determining the quality of the operation status of the production line, the device states of a plurality of the production devices including identification information for identifying abnormal states obtained by subdividing the device states of the production devices, and the processing status of the work processed by the production device, a data acquisition unit that acquires; Using a learned model for inferring the degree of influence on the decline in the key performance evaluation index of the stop factor in the key performance evaluation index of the production line, the device states of a plurality of the production devices, and the processing status of the work processed by the production device, an inference unit that outputs the degree of influence on the decline in the key performance evaluation index of the stop factor from the key performance evaluation index, the device states of a plurality of the production devices, and the processing status of the work acquired by the data acquisition unit; comprising; The learned model individually outputs the degree of influence on the decline in the key performance evaluation index of the stop factor for a plurality of the stop factors; An inference device characterized by this.

14. For a production system that analyzes the stop factors of a production line including a plurality of production devices, the key performance evaluation index that is the criterion for determining the quality of the operation status of the production line, the device states of a plurality of the production devices including identification information for identifying abnormal states obtained by subdividing the device states of the production devices, the processing status of the work processed by the production device, and the degree of influence on the decline in the key performance evaluation index of the stop factor in the key performance evaluation index, the device states of a plurality of the production devices, and the processing status of the work, is generated by performing machine learning using learning data including; The degree of influence on the decline in the key performance evaluation index of the stop factor is individually output for a plurality of the stop factors from the key performance evaluation index, the device states of a plurality of the production devices, and the processing status of the work; A learned model characterized by this.

15. From a production system that analyzes the causes of stoppage of a production line including a plurality of production devices, an important performance evaluation index that is a criterion for determining whether the operation status of the production line is good or bad, and identification information for identifying abnormal states obtained by subdividing the device status of the production devices, the device status of the plurality of production devices, the processing status of the work processed by the production devices, and the degree of influence on the decrease in the important performance evaluation index of the stoppage cause in the important performance evaluation index, the device status of the plurality of production devices, and the processing status of the work. A data acquisition step of acquiring learning data including: A model generation step of generating a learned model for inferring the degree of influence on the decrease in the important performance evaluation index of the stoppage cause from the important performance evaluation index of the production line, the device status of the plurality of production devices, and the processing status of the work processed by the production devices using the learning data; including The learned model individually outputs the degree of influence on the decrease in the important performance evaluation index of the stoppage cause for the plurality of stoppage causes. A method for generating a learned model, characterized by the above.

16. By performing the data acquisition step of acquiring new learning data and performing the model generation step using the new learning data, the learned model is updated. The method for generating a learned model according to claim 15, characterized by the above.

Images