Abnormality detection device, abnormality detection method, and program

Abstract

The abnormality detection method includes: acquiring a standard value per unit time of a plurality of variables indicating a condition of a production line when the production line operates normally; collecting live values of the plurality of variables per unit time; calculating, for each unit time, a statistic representing the degree of deviation of the live values of the plurality of variables with respect to the standard values of the plurality of variables; and detecting that an abnormality has occurred in the production line on the basis of the condition that the statistical magnitude is outside the reference range.

Description

Technical Field

[0001] The present disclosure relates to an abnormality detection device, an abnormality detection method, and a program. Background Art

[0002] In order to improve production efficiency, it is desirable to detect abnormalities in the production line in advance. Japanese Patent Publication No. 2019-177783 (Patent Document 1) discloses a technology for detecting the deterioration of the ride comfort of a railway vehicle based on the increase in the Mahalanobis distance between characteristic data obtained from an acceleration sensor installed in a railway vehicle and pre-acquired unit data.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Publication No. 2019-177783

[0006] Patent Document 2: Japanese Patent Application Publication No. 2003-44115

[0007] Non-patent literature

[0008] Non-patent document 1: Tamura Kishiomi, "The fifth MT system capable of direction determination - TS method, T method", Standardization and Quality Management, 2009, Vol. 62, No. 2

[0009] Non-patent document 2: Genichi Taguchi, "Target Function and Basic Function (6) - Comprehensive Prediction Based on T Method -", Quality Engineering, 2005, Vol. 13, No. 3, p. 5-10

[0010] Non-patent document 3: "Concepts and application of quality engineering in the development and design stage - System evaluation and improvement without trial production / testing -", [online], [retrieved on March 22, 2023], Internet <https: / / foundry.jp / bukai / wp-content / uploads / 2012 / 07 / e4806f10b0797ec0932d9317dd92a533.pdf> Summary of the invention

[0011] Problems to be solved by the invention

[0012] The production line includes a variety of equipment and a large number of variables can be obtained. Therefore, when the technology disclosed in Japanese Patent Publication No. 2019-177783 (Patent Document 1) is applied to the production line, in order to confirm whether the production line has abnormalities, it is necessary to compare the benchmark and the measured values ​​of a large number of variables. As a result, the number of operations to confirm whether the production line has abnormalities increases.

[0013] The present disclosure has been made in view of the above-mentioned problems, and an object of the present disclosure is to provide an abnormality detection device, an abnormality detection method, and a program that can easily confirm whether or not a monitoring object in a production site has an abnormality.

[0014] Means for solving problems

[0015] According to an example of the present disclosure, an abnormality detection device has an acquisition unit, a collection unit, a calculation unit, and a detection unit. The acquisition unit acquires standard values ​​of multiple variables representing the status of the monitored object in a unit interval when the monitored object in the production site is operating normally. The collection unit collects actual values ​​of multiple variables in the unit interval. The calculation unit calculates a statistic representing the deviation of the actual values ​​of multiple variables relative to the standard values ​​of multiple variables. The detection unit detects that an abnormality has occurred in the monitored object based on the statistic being outside the reference range.

[0016] According to the above disclosure, the user can easily confirm whether or not there is an abnormality in the monitored object in the production site by confirming the result of the detection unit.

[0017] In the above disclosure, the abnormality detection device further includes a selection unit that selects a valid variable having a relatively large influence on a statistic from among a plurality of variables.

[0018] According to the above disclosure, the user can confirm the selection result of the selection unit and determine the effective variable as the candidate cause of the abnormality of the monitored object. As a result, the user can take measures such as maintenance of the monitored object or re-evaluate the production plan in advance to adjust the effective variable.

[0019] In the above disclosure, the selection unit selects, as the above effective variable, a variable that is effective for improving the estimation accuracy of a T-method estimation model that uses a plurality of variables as explanatory variables and a statistic as a target variable.

[0020] According to the above disclosure, it is possible to easily select effective variables having a relatively large influence on statistics.

[0021] In the above disclosure, the abnormality detection device further includes an analysis unit that performs a principal component analysis using actual values ​​and statistics of effective variables for each unit interval.

[0022] According to the above disclosure, users can select variables with high influence on statistics from multiple valid variables based on the principal component loadings obtained through principal component analysis. As a result, users can perform production line maintenance, re-evaluate production plans, and other countermeasures in advance to adjust the selected variables.

[0023] In the above disclosure, the analysis unit selects two principal components that maximize the variance of the statistic, and determines the influence of the effective variable on the statistic based on the principal component loadings of the effective variable and the statistic with respect to the two principal components.

[0024] According to the above disclosure, the user can select effective variables with high influence on statistics from among multiple effective variables based on the determined influence. As a result, the user can perform measures such as maintenance of the monitored object or re-evaluation of the production plan in advance to adjust the selected effective variables.

[0025] In the above disclosure, the acquisition unit uses a virtual model of the monitored object to perform simulation of the values ​​of a plurality of variables when the monitored object is operated according to the production plan, and the acquisition unit acquires the value of the unit interval obtained by the simulation as the standard value.

[0026] Alternatively, in the above disclosure, the acquisition unit may acquire, as the standard values, values ​​of a plurality of variables in a unit section when the monitored object was operating normally in the past.

[0027] In the above disclosure, the statistic is, for example, the standard SN ratio. Thus, the statistic properly represents the similarity between the actual value and the standard value.

[0028] In the above disclosure, the monitoring object is the production line, and the unit section is the unit time. According to the present disclosure, the user can easily check whether there is an abnormality in the production line by checking the result of the detection unit.

[0029] In the above disclosure, the monitoring object includes the manufacturing device. According to the present disclosure, the user can easily check whether there is an abnormality in the manufacturing device by checking the result of the detection unit.

[0030] In the above disclosure, the manufacturing device repeatedly performs the object action. The unit interval is the period during which the object action is performed. According to the present disclosure, a plurality of variables represent the status of the manufacturing device during the period during which the object action is performed. Therefore, the user can simply confirm whether there is an abnormality for each object action repeatedly performed by the manufacturing device. According to another example of the present disclosure, the abnormality detection method has steps 1 to 4. The first step is the following step: obtaining the standard values ​​of a plurality of variables representing the status of the monitored object in the unit interval when the monitored object is operating normally. The second step is the following step: collecting the actual values ​​of the plurality of variables in the unit interval. The third step is the following step: calculating a statistic representing the deviation of the actual values ​​of the plurality of variables relative to the standard values ​​of the plurality of variables. The fourth step is the following step: detecting that an abnormality has occurred in the monitored object based on the statistic being outside the reference range. According to another example of the present disclosure, the program causes the computer to execute the above-mentioned abnormality detection method. According to these disclosures, the user can simply confirm whether there is an abnormality in the monitored object in the production site.

[0031] Effects of the Invention

[0032] According to the present disclosure, a user can easily check whether a monitoring object in a production site has abnormality. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a schematic diagram showing the configuration of a system including an abnormality detection device according to an embodiment.

[0034] Figure 2 This is a diagram showing an example of a production line.

[0035] Figure 3 It is a schematic diagram showing an example of the hardware configuration of the abnormality detection device according to the embodiment.

[0036] Figure 4 This is a diagram showing an example of the functional structure of the abnormality detection device according to the embodiment.

[0037] Figure 5 This is a diagram for explaining the processing of the simulation execution unit.

[0038] Figure 6 This is a diagram for explaining the processing of the calculation unit.

[0039] Figure 7 This is a diagram showing an example of the selection result of the effective variable.

[0040] Figure 8 This is a diagram showing an example of the analysis result of the analysis unit.

[0041] Fig. 9 This is a flowchart showing an example of the processing flow of the abnormality detection device.

[0042] Fig.10 This is a diagram showing the functional structure of the abnormality detection device according to Modification Example 1.

[0043] Fig.11 This is a diagram showing an example of a manufacturing apparatus to be monitored.

[0044] Fig.12 This is a diagram showing the transition of parameter values ​​during the execution of the injection molding operation.

[0045] Fig.13 It is a diagram showing standard values ​​and actual values ​​of a plurality of variables indicating the status of the injection molding machine 3 .

[0046] Fig.14 This is a diagram showing the transition of the statistical quantity calculated for each injection molding operation.

[0047] Fig.15 This is a diagram showing the results of principal component analysis using actual values ​​of effective variables and standard SN ratios for each injection molding period.

[0048] Fig.16 This is a diagram showing another example of a manufacturing apparatus to be monitored.

[0049] Fig.17 This is a diagram showing the transition of the statistics calculated for each joining operation. DETAILED DESCRIPTION

[0050] Embodiments of the present invention will be described in detail with reference to the accompanying drawings. In addition, the same or corresponding parts in the drawings are denoted by the same reference numerals and their description will not be repeated.

[0051] §1 Application Examples

[0052] Reference Figure 1 , an example of a scenario in which the present invention is applied is described. Figure 1 1 is a schematic diagram showing the structure of a system including an abnormality detection device according to an embodiment. Figure 1 As shown, the system 1 includes a production line 2 as an example of a monitoring target in a production site, and an abnormality detection device 100 for detecting abnormality in the production line 2. The abnormality detection device 100 is communicatively connected to the production line 2. The monitoring target in the production site is not limited to the production line 2.

[0053] The abnormality detection device 100 obtains standard values ​​per unit time of multiple variables representing the status of production line 2 when production line 2 is operating normally (step (1)). The multiple variables include the number of devices, the number of personnel, the output, etc. of each process constituting production line 2. The standard values ​​of the multiple variables represent the standard status of production line 2. The unit time is, for example, 5 minutes. In addition, the standard values ​​of the multiple variables may be different for each unit time. For example, the standard value within a unit time just after the operation of production line 2 begins may be different from the standard value within a unit time after a few hours from the start of operation of production line 2. Considering such a situation, the abnormality detection device 100 obtains the standard value for each unit time.

[0054] Next, the abnormality detection device 100 collects the actual values ​​of the multiple variables per unit time when the operation of the production line 2 starts (step (2)). Then, the abnormality detection device 100 calculates the statistics representing the deviation of the actual values ​​of the multiple variables from the standard values ​​of the multiple variables for each unit time (step (3)). The abnormality detection device 100 detects that an abnormality has occurred in the production line 2 when the statistics are outside the reference range (step (4)).

[0055] Through step (3), even if the number of multiple variables representing the status of production line 2 is large, the deviation of the current status of production line 2 from the standard status is represented by a statistic. As a statistic, for example, the standard SN ratio can be used. However, the statistic is not limited to the standard SN ratio, as long as it is a parameter that represents the deviation between the standard value and the actual value of multiple variables. The greater the deviation between the standard value and the actual value of multiple variables, the smaller the standard SN ratio. Therefore, a decrease in the standard SN ratio means that the actual values ​​of multiple variables deviate from the standard values. The threshold TH is predetermined in such a way that it becomes the value when production line 2 changes from normal to abnormal. In this case, the reference range is the range above the threshold TH. Therefore, the standard SN ratio outside the reference range (i.e., less than the threshold TH) means that some kind of abnormality has occurred in production line 2. Therefore, by confirming the result of the above step (4), the user can simply confirm whether there is any abnormality in production line 2.

[0056] §2Concrete examples

[0057] <Production Line Example>

[0058] Figure 2 This is a diagram showing an example of a production line. Figure 2 The production line 2 shown includes a cutting process 21 , a component shipping process 22 , an assembly process 23 , and a product shipping process 24 .

[0059] The cutting process 21 is to produce a metal member of a predetermined shape by cutting a metal sheet, for example. The cutting process 21 includes one or more cutting devices 21a, a timer 21b, and counters 21c and 21d.

[0060] The one or more cutting devices 21a output a first signal indicating whether they are operating normally, a second signal indicating whether the cutting operation has been performed, and a third signal indicating that the production of the metal component has been completed.

[0061] The counter 21c measures the number of cutting devices 21a operating normally in the cutting process 21 (hereinafter referred to as the “number of devices”) based on the first signal.

[0062] The timer 21b measures the time during which the cutting operation is performed per unit time (hereinafter referred to as "cutting time") based on the second signal.

[0063] The counter 21d measures the number of metal parts produced per unit time (hereinafter referred to as "throughput (cutting process)") based on the third signal.

[0064] The component delivery process 22 transports the assembly components purchased from the outside to the assembly process 23. The component delivery process 22 includes a sensor 22a and an automatic guided vehicle (AGV) 22b. In the component delivery process 22, after the operator performs the operation of taking out the assembly components purchased from the outside from the package, the automatic guided vehicle 22b transports the assembly components.

[0065] The sensor 22a measures the number of workers per unit time (hereinafter referred to as "number of workers (parts shipping process)"). The sensor 22a is, for example, an image sensor, and measures the number of workers (parts shipping process) using an image obtained by photographing the parts shipping process 22 .

[0066] The automated guided vehicle 22b has a timer 22c and a counter 22d. The timer 22c measures the time for transporting assembly parts per unit time (hereinafter referred to as "AGV operation time"). The counter 22d measures the number of assembly parts transported per unit time (hereinafter referred to as "output (parts shipping process)").

[0067] In the assembly process 23, the assembly components conveyed from the component delivery process 22 are assembled to the metal components produced in the cutting process 21, thereby manufacturing the product. In the assembly process 23, for example, an operator performs the assembly work and places the completed product on a belt conveyor.

[0068] The assembly process 23 is provided with a sensor 23a, which measures the number of metal parts before assembly of the assembly parts (hereinafter referred to as "the number of work-in-progress") among the metal parts conveyed from the cutting process 21 at unit time. The sensor 23a is, for example, an image sensor, and measures the number of work-in-progress using an image obtained by photographing a tray on which the metal parts before assembly of the assembly parts are placed.

[0069] Furthermore, the assembly process 23 is provided with: a sensor 23b that measures the number of operators (hereinafter referred to as "number of operators (assembly process)") per unit time; and a counter 23c that measures the number of products manufactured per unit time (hereinafter referred to as "output (assembly process)"). The sensor 23b is, for example, an image sensor that measures the number of operators (assembly process) using an image obtained by photographing the assembly process 23. The counter 23c is, for example, a limit switch provided on a belt conveyor on which the product is placed. In the product shipping process 24, the product is packaged and shipped.

[0070] <Hardware Structure of Abnormality Detection Device>

[0071] Figure 3Schematic diagram showing an example of the hardware structure of the abnormality detection device according to the embodiment. Figure 3 As shown, the abnormality detection device 100 typically has a structure that complies with a general computer architecture. Specifically, the abnormality detection device 100 includes a processor 101 such as a CPU (Central Processing Unit) or an MPU (Micro-Processing Unit), a memory 102, a storage 103, a display controller 104, an input interface 105, and a communication interface 106. These various parts are connected via a bus in a manner that allows data communication with each other.

[0072] The processor 101 implements various processes of the present embodiment by developing various programs stored in the storage 103 in the memory 102 and executing the programs.

[0073] The memory 102 is typically a volatile storage device such as a DRAM (Dynamic Random Access Memory), and stores programs and the like read from the storage 103 .

[0074] The memory 103 is typically a nonvolatile magnetic storage device such as a hard disk drive. The memory 103 stores an abnormality detection program 131 and a simulation program 132 executed by the processor 101. Various programs installed in the memory 103 are distributed in a state of being stored in a memory card or the like.

[0075] The display controller 104 is connected to the display device 70 , and outputs a signal for displaying various information to the display device 70 in accordance with an internal command from the processor 101 .

[0076] The input interface 105 mediates data transmission between the processor 101 and an input device 75 such as a keyboard, a mouse, a touch panel, or a dedicated console. That is, the input interface 105 receives an operation instruction given by a user operating the input device 75 .

[0077] The communication interface 106 mediates data transmission between the processor 101 and external devices (e.g., various devices installed in the production line 2). The communication interface 106 typically includes Ethernet (registered trademark), USB (Universal Serial Bus), etc. In addition, various programs stored in the memory 103 can also be downloaded from a distribution server, etc. via the communication interface 106.

[0078] When using a computer having a structure conforming to a general computer architecture as described above, in addition to the application for providing the functions of the present embodiment, an OS (Operating System) for providing the basic functions of the computer may be installed. In this case, the program of the present embodiment may also call necessary modules among the program modules provided as part of the OS in a prescribed order and timing to execute processing. That is, the program of the present embodiment itself may sometimes not include the above-mentioned modules, but cooperate with the OS to execute processing.

[0079] Alternatively, part or all of the functions provided by the execution of the abnormality detection program 131 and the simulation program 132 may be implemented as a dedicated hardware circuit.

[0080] <Functional Structure of Abnormality Detection Device>

[0081] Figure 4 FIG. 1 is a diagram showing an example of the functional structure of an abnormality detection device according to an embodiment. Figure 4 As shown, the abnormality detection device 100 includes a simulation execution unit 11, a collection unit 12, a calculation unit 13, a detection unit 14, a selection unit 15, and an analysis unit 16. The simulation execution unit 11 is executed by a processor 101 (see Figure 3 ) executes the simulation program 132. The collection unit 12 is implemented by the communication interface 106 and the processor 101 executing the abnormality detection program 131. The calculation unit 13, the detection unit 14, the selection unit 15 and the analysis unit 16 are implemented by the processor 101 executing the abnormality detection program 131.

[0082] The simulation execution unit 11 operates as an acquisition unit that acquires standard values ​​of a plurality of variables that represent the status of the production line 2 per unit time when the production line 2 is operating normally. Specifically, the simulation execution unit 11 simulates the values ​​per unit time (e.g., 5 minutes) of a plurality of variables that represent the status of the production line 2 by inputting production plan data that represent the production plan into the virtual model of the production line 2. The simulation execution unit 11 may perform the simulation using a known technique (e.g., Japanese Patent Application Laid-Open No. 2003-44115 (Patent Document 2)).

[0083] Figure 5 2M is a diagram for explaining the processing of the simulation execution unit. The virtual model 2M is pre-created based on the production line 2. The virtual model 2M outputs the values ​​per unit time of a plurality of variables representing the status of the production line 2 when the production line 2 is operated according to the production plan, based on the input of various data representing the production plan. Figure 5As shown, multiple variables representing the status of production line 2 include "cutting time", "number of devices", "output (cutting process)", "AGV operation time", "number of personnel (parts shipping process)", "output (parts shipping process)", "number of work-in-progress", "number of personnel (assembly process)", and "output (assembly process)".

[0084] The virtual model 2M includes a first model 21M corresponding to the cutting process 21, a second model 22M corresponding to the component delivery process 22, and a third model 23M corresponding to the assembly process 23. The first model 21M is created according to the specification of the cutting device 21a. The first model 21M is defined by the following parameters.

[0085] The cutting speed of the cutting device 21a during normal operation;

[0086] The amount of cutting required to make one metal part;

[0087] · Preparation time from the completion of cutting of a metal part to the start of cutting of the next metal part, etc. The first model 21M receives input of production plan data indicating the number of cutting devices 21a in operation, the scheduled operation period, the stop period (rest time), etc., and outputs the values ​​of the cutting process time, the number of devices, and the output per unit time of the cutting process 21.

[0088] The second model 22M is created based on the specifications of the automated guided vehicle 22b and the specifications of the standard work of the operator in the component shipping process 22. The second model 22M is defined by the following parameters.

[0089] The time required for an operator to remove an assembly part from its packaging when performing standard operations;

[0090] The time required for the operator to place the assembly parts on the automatic transport vehicle 22b;

[0091] The time required for the automatic transport vehicle 22b to transport one assembly component to the assembly process 23, etc.

[0092] The second model 22M receives input of production plan data indicating the number of workers assigned to the component shipping process 22, the scheduled operation period, the workers' rest time, etc., and outputs the values ​​of the AGV operation time, the number of workers, and the production volume of the component shipping process 22 per unit time.

[0093] The third model 23M is created based on the specification of the standard operation of the operator. The third model 23M is defined, for example, by the time required for the operator to complete the standard operation of one product. The third model 23M receives the production plan data indicating the number of operators assigned to the assembly process 23, the scheduled operation period, the rest time, etc., the output data of the first model 21M, and the output data of the second model 22M, and outputs the value of the number of work-in-progress, the number of workers, and the output of the assembly process 23 per unit time.

[0094] The virtual model 2M is a model based on the premise that the devices and operators perform standard operations in each process. Therefore, the simulation execution unit 11 obtains the values ​​of multiple variables per unit time (e.g., 1 minute, 5 minutes, etc.) output from the virtual model 2M as the values ​​(standard values) of the multiple variables when the production line 2 is operating normally.

[0095] Figure 4 The collection unit 12 shown collects the actual values ​​of multiple variables representing the current status of the production line 2 per unit time (for example, 1 minute, 5 minutes, etc.) from the production line 2. Specifically, the collection unit 12 collects the actual values ​​per unit time of "cutting time", "number of devices", and "output (cutting process)" from the timer 21b, counter 21c, and counter 21d of the cutting process 21. The collection unit 12 collects the actual values ​​per unit time of "AGV operation time", "number of personnel (parts shipping process)", and "output (parts shipping process)" from the timer 22c, sensor 22a, and counter 22d of the component shipping process 22. The collection unit 12 collects the actual values ​​per unit time of "number of products in process", "number of personnel (assembly process)", and "output (assembly process)" from the sensor 23a, sensor 23b, and counter 23c of the assembly process 23.

[0096] The calculation unit 13 calculates the standard SN ratio as a statistic representing the degree of deviation of the actual values ​​of the plurality of variables relative to the standard values ​​of the plurality of variables for each unit time. The standard SN ratio is based on "The concept and flexible application of quality engineering in the development and design stage - System evaluation and improvement without trial production / testing -", [online], [retrieved on January 4, 2022], Internet<https: / / foundry.jp / bukai / wp-content / uploads / 2012 / 07 / e4806f10b0797ec0932d9317dd92a533.pdf> ” (Non-patent document 3) calculated.

[0097] Figure 6 is a diagram for explaining the processing of the calculation unit. Figure 6, an example of using "cutting time", "number of devices", "output (cutting process)", "AGV operation time", "number of personnel (parts shipping process)", "output (parts shipping process)", "number of products in process", "number of personnel (assembly process)", and "output (assembly process)" as multiple variables is shown. The calculation unit 13 calculates the standard SN ratio of the actual values ​​of the multiple variables collected by the collection unit 12 relative to the standard values ​​of the multiple variables calculated by the simulation execution unit 11 for each unit time. The calculation unit 13 can also display the graph 30 showing the transition of the calculated standard SN ratio on the display device 70. Thus, the user can understand the transition of the status of the production line 2 by confirming the graph 30.

[0098] Figure 4 The detection unit 14 shown uses the standard SN ratio calculated by the calculation unit 13 to detect that an abnormality has occurred in the production line 2. The greater the deviation between the standard value and the actual value of the plurality of variables, the smaller the standard SN ratio. Therefore, the detection unit 14 detects that an abnormality has occurred in the production line 2 based on the fact that the standard SN ratio is outside the reference range (i.e., less than the threshold value TH). For example, Figure 6 In the case of the graph 30 shown in FIG. 1 , the detection unit 14 detects that an abnormality has occurred in the production line 2 at the timing T when the standard SN ratio is less than the threshold value TH. The detection unit 14 may also display an error notification on the display device 70 based on the detection of the abnormality in the production line 2. Thus, the user can reliably understand that an abnormality has occurred in the production line 2.

[0099] Figure 4 The selection unit 15 shown selects an effective variable having a relatively large influence on the standard SN ratio from a plurality of variables according to the occurrence of an abnormality detected by the detection unit 14. Specifically, the selection unit 15 selects a variable effective for improving the estimation accuracy of the T-method estimation model using the plurality of variables as explanatory variables and the standard SN ratio as the target variable as an effective variable. Thus, the user can select an effective variable as a candidate cause of the abnormality of the production line 2, thereby enabling the abnormality of the production line 2 to be eliminated in advance. In addition, the selection unit 15 may select a plurality of effective variables.

[0100] The T method is one of the multivariate analysis methods provided by statistician Genichi Taguchi. It is a method of estimating the value of a target variable based on the values ​​of multiple explanatory variables (see "Tamura Kishishin, "The fifth MT system capable of direction determination - TS method, T method", Standardization and Quality Management, 2009, Vol. 62, No. 2" (Non-patent document 1)).

[0101] The selection unit 15 generates a T-method estimation model by performing T-method data processing on a data set representing actual values ​​of multiple variables per unit time and values ​​of the standard SN ratio. In the T-method data processing, a univariate regression is performed for each explanatory variable, and a weighted average is performed for each explanatory variable, thereby generating a T-method estimation model for calculating an estimated value of the target variable. The estimation model obtained by the T-method data processing is represented by the following formula (1).

[0102]

[0103] In formula (1), Xij is the standardized value of the jth explanatory variable in the i-th unit time. Standardization is the process of subtracting the average value of the variable represented by the data set. ηj represents the SN ratio of the j-th explanatory variable. The SN ratio ηj represents the linearity between the value of the j-th explanatory variable and the value of the target variable, indicating the estimated accuracy of the value of the target variable. βj represents the proportionality constant of the univariate regression. Mi is the estimated value of the target variable in the i-th unit time after standardized processing.

[0104] For example, the selection unit 15 uses the orthogonal table of the two-level system to select effective variables that are effective in improving the estimation accuracy of the target variable from multiple variables. The selection unit 15 uses, for example, the method described in "Genichi Taguchi, "Target Function and Basic Function (6) - Comprehensive Prediction Based on T Method -", Quality Engineering, 2005, Vol. 13, No. 3, p. 5-10" (Non-Patent Document 2) to select effective variables. By using the selection of effective variables using the orthogonal table of the two-level system (also called "item selection"), it is possible to select explanatory variables that are effective in improving the estimation accuracy of the target variable for each explanatory variable (also called "item"), and exclude explanatory variables that have an adverse effect on the estimation accuracy of the target variable (reduce the estimation accuracy). Furthermore, by using the orthogonal table of the two-level system, it is possible to consider the impact when there is interaction between items. Specifically, according to the method described in "Genichi Taguchi, "Target Function and Basic Function (6) - Comprehensive Prediction Based on T Method -", Quality Engineering, 2005, Vol. 13, No. 3, p. 5-10" (Non-Patent Literature 2), an orthogonal table is created to which one of two levels of use or non-use is assigned for each of multiple explanatory variables (items), and a comprehensive SN ratio is calculated for each row of the orthogonal table. The selection unit 15 determines the estimated effect of the target variable of the item and whether there is any adverse effect on the estimation of the target variable based on the comprehensive SN ratio when each item is used and when each item is not used. The selection unit 15 selects a valid variable from multiple variables based on the judgment result.

[0105] Alternatively, the selection unit 15 may generate a plurality of patterns representing combinations of two or more variables among the plurality of variables instead of using the orthogonal table of the two-level system. Then, the selection unit 15 may generate a T-method estimation model using the actual values ​​of the two or more variables represented by each pattern as explanatory variables and the standard SN ratio as the target variable, and select effective variables based on the size of the comprehensive SN ratio of the generated T-method estimation model. For example, the selection unit 15 selects the variable included in the pattern with the largest comprehensive SN ratio as the effective variable.

[0106] Figure 7 is a diagram showing an example of the selection results of effective variables. Figure 7 In the example shown, "AGC operating time", "number of devices", "number of personnel (parts shipment)", and "processing time" are selected as valid variables. Figure 7 The vertical axis of represents the SN ratio ηj of each variable in the T method estimation model using all the multiple variables. As described above, the SN ratio ηj of each variable represents the magnitude of the influence on the standard SN ratio as the target variable.

[0107] Figure 4 The analysis unit 16 shown performs principal component analysis (PCA) using actual values ​​of effective variables per unit time and standard SN ratios.

[0108] Figure 8 This is a diagram showing an example of the analysis result of the analysis unit. Figure 8 The results of principal component analysis are shown for "AGC operation time", "number of devices", "number of personnel (parts shipment)", "processing time", and "standard SN ratio" per unit time.

[0109] The analyzing unit 16 selects two principal components that maximize the variance of the standard SN ratio. Specifically, the analyzing unit 16 selects the first principal component having the largest contribution rate and the second principal component having the second largest contribution rate.

[0110] Alternatively, the analysis unit 16 may display a scatter diagram of each combination of two principal components among the plurality of principal components obtained by the principal component analysis on the display device 70, and select two principal components of the combination corresponding to the scatter diagram specified by the user. In the scatter diagram, points corresponding to the principal component scores of each of the two principal components obtained from the values ​​of "AGC operation time", "number of devices", "number of personnel (parts shipment)", "processing time" and "standard SN ratio" per unit time are plotted. The analysis unit 16 may also make the display form of each point in the scatter diagram different according to the value of the standard SN ratio corresponding to the point. For example, the analysis unit 16 displays the points whose values ​​of the standard SN ratio are within the first range (low range) in black, displays the points whose values ​​of the standard SN ratio are within the second range (medium range) larger than the first range in gray, and displays the points whose values ​​of the standard SN ratio are within the third range (high range) larger than the second range in white. Thus, the user confirms the scatter diagram of each combination and selects the scatter diagram with the most separated values ​​of the standard SN ratio.

[0111] like Figure 8 As shown, the analysis unit 16 displays a biplot 40 of the selected two principal components on the display device 70. In the biplot 40, the principal component scores and the principal component loads are displayed in an overlapping manner. As described above, the display form of the points corresponding to the principal component scores of the two principal components obtained from the data per unit time is different depending on the value of the standard SN ratio. Therefore, it can be understood that the size of the standard SN ratio is separated by using the selected two principal components.

[0112] The principal component loading indicates the correlation between the variable value and the principal component score. The larger the principal component loading, the stronger the correlation between the principal component and the variable. Figure 8 As shown, the principal component loading of each variable is represented by a vector. Vector v0 represents the principal component loading of "standard SN ratio". Vector v1 represents the principal component loading of "AGC operation time". Vector v2 represents the principal component loading of "number of devices". Vector v3 represents the principal component loading of "number of personnel (parts shipment)". Vector v4 represents the principal component loading of "cutting processing time".

[0113] The user can understand the variables that affect the reduction of the standard SN ratio by checking the biplot 40. That is, the user can identify the variables corresponding to the vectors with large components parallel to the vector v0 among the vectors v1 to v4 as the variables that affect the reduction of the standard SN ratio. Figure 8 In the biplot 40 shown, the user can identify the variables “AGV operation time” and “cutting time” corresponding to the vectors v1 and v4 that are not perpendicular to the vector v0 and have large magnitudes as variables that affect the reduction of the standard SN ratio.

[0114] The variable that affects the decrease in the standard SN ratio is considered to be the cause of the abnormality of the production line 2. Therefore, the user can eliminate the abnormality of the production line 2 by adjusting the variable that affects the decrease in the standard SN ratio.

[0115] In addition, the analysis unit 16 may determine the influence of the effective variable on the standard SN ratio based on the principal component loadings of the effective variable and the standard SN ratio with respect to the two selected principal components, and display the determined influence on the display device 70. Specifically, the analysis unit 16 calculates the vector ( ) representing the principal component loading of each effective variable for each of the plurality of effective variables. Figure 8 v1~v4) and the vector representing the principal component loading of the “standard SN ratio” ( Figure 8 The absolute value of the inner product of the vector v0 of the standard SN ratio is used as the influence on the standard SN ratio. Thus, the user can easily identify the variable that affects the reduction of the standard SN ratio by confirming the influence displayed on the display device 70.

[0116] <Processing flow of abnormality detection device>

[0117] Fig. 9 FIG. 1 is a flowchart showing an example of the processing flow of the abnormality detection device. Fig. 9 As shown, first, the processor 101 of the abnormality detection device 100 obtains the standard values ​​per unit time of multiple variables representing the status of the production line 2 when the production line 2 is operating normally (step S1). In this embodiment, the processor 101 inputs the production plan data into the virtual model 2M, and obtains the values ​​of the multiple variables per unit time output from the virtual model 2M as the above-mentioned standard values.

[0118] Next, the processor 101 collects actual values ​​of multiple variables per unit time from the production line 2 (step S2). The processor 101 calculates a standard SN ratio representing the degree of deviation of the actual values ​​of the multiple variables relative to the standard values ​​of the multiple variables for each unit time (step S3). The processor 101 displays the graph 30 representing the transition of the standard SN ratio on the display device 70 (step S4).

[0119] Next, the processor 101 determines whether the standard SN ratio is outside the reference range (ie, less than the threshold TH) (step S5). If the standard SN ratio is within the reference range (ie, greater than the threshold TH) (step S5: No), the process returns to step S2.

[0120] When the standard SN ratio is outside the reference range (step S5: Yes), the processor 101 notifies that an abnormality has occurred in the production line 2 (step S6). For example, the processor 101 displays an error notification screen on the display device 70.

[0121] Next, the processor 101 selects effective variables that are effective for improving the estimation accuracy of the T-method estimation model that uses the actual values ​​of the multiple variables as explanatory variables and the standard SN ratio as the target variable from the multiple variables (step S7). The processor 101 may also display a screen showing the selection result on the display device 70. The effective variables are items that are important for the estimation accuracy of the standard SN ratio. Therefore, by confirming the selection result, the user can determine the effective variables as the cause candidate of the abnormality of the production line 2. As a result, the user can perform maintenance of the production line 2 to adjust the effective variables and re-evaluate the production plan.

[0122] Next, the processor 101 performs a principal component analysis using the actual value of the effective variable per unit time and the standard SN ratio (step S8). The processor 101 displays the result of the principal component analysis on the display device 70 (step S9). For example, the processor 101 displays Figure 8 The biplot 40 shown is displayed on the display device 70. Alternatively, the processor 101 may determine the influence of the effective variable on the standard SN ratio based on the principal component loading of each of the effective variable and the standard SN ratio, and display the determined influence on the display device 70. Thus, the user can filter out the variables with high influence on the standard SN ratio among the multiple effective variables. As a result, the user can perform maintenance of the production line 2 by adjusting the filtered variables. For example, the user performs maintenance of the cutting device 21a of the cutting process 21. Alternatively, the user may re-evaluate the production plan. For example, the user can adjust the number of operators configured in the assembly process 23, adjust the number of cutting devices 21a operating in the cutting process 21, etc.

[0123] Next, the processor 101 determines whether the production plan has been re-evaluated (step S10 ). Specifically, the processor 101 determines that the production plan has been re-evaluated based on the fact that new production plan data has been input.

[0124] If the production plan is re-evaluated (step S10: Yes), the process returns to step S1. As a result, the standard values ​​per unit time of the plurality of variables representing the status of the production line are updated. If the production plan is not re-evaluated (step S10: No), the process returns to step S2.

[0125] <Modification 1>

[0126] Fig.10 This is a diagram showing the functional structure of the abnormality detection device according to Modification Example 1. Fig.10 The abnormality detection device 100A shown has the same hardware structure as the abnormality detection device 100 (see Figure 3 ).like Fig.10 As shown, the abnormality detection device 100A and Figure 4 The abnormality detection device 100 shown in the figure is different in that it has a data set selection unit 17 and a real-time database 18 instead of the simulation execution unit 11. The data set selection unit 17 is implemented by the processor 101 executing the abnormality detection program 131. The real-time database 18 is implemented by the memory 103.

[0127] The actual situation database 18 includes one or more data sets indicating values ​​per unit time of a plurality of variables when the production line 2 was operating normally in the past.

[0128] When the operating conditions of the production line 2 are variable, information indicating the operating conditions of the production line 2 may be added to each data set. The operating conditions include the number of cutting devices 21a operating in the cutting process 21, the number of operators configured in the component shipping process 22, the number of operators configured in the assembly process 23, and the like.

[0129] The data set selection unit 17 operates as an acquisition unit that acquires standard values ​​of a plurality of variables representing the status of the production line 2 per unit time when the production line 2 is operating normally. Specifically, the data set selection unit 17 selects one or more data sets from the actual database 18 according to the user's instruction. For example, the user may specify one or more data sets when the production line 2 is operating in an ideal state from the actual database 18. Alternatively, when information representing the operating conditions is attached to each data set, the user may specify one or more data sets to which information representing the operating conditions consistent with the production plan is attached.

[0130] The data set selection unit 17 obtains the values ​​of a plurality of variables per unit time represented by the selected one or more data sets as standard values. When one data set is selected, the data set selection unit 17 obtains the value represented by the data set as the standard value. When a plurality of data sets are selected, the data set selection unit 17 obtains, for each of the plurality of variables, a representative value (e.g., an average value, a median value, etc.) of the values ​​per unit time represented by the plurality of data sets as the standard value.

[0131] According to the first modification, the data set selection unit 17 obtains the values ​​per unit time of the plurality of variables when the production line 2 was operating normally in the past as standard values. Thus, the standard SN ratio calculated by the calculation unit 13 indicates the degree of deviation between the status of the production line 2 when operating normally and the current status. Therefore, the fact that the standard SN ratio is less than the threshold value means that some abnormality has occurred in the production line 2. As a result, the user can easily confirm whether there is an abnormality in the production line by confirming the detection result of the detection unit 14.

[0132] The processing flow of the abnormality detection device 100A in the modification example 1 is the same as Fig. 9If the data set does not contain information indicating the operating conditions, step S10 is omitted and the process returns to step S1 after step S9.

[0133] <Modification 2>

[0134] exist Figure 1 In the embodiment shown, the monitoring object of the production site is production line 2. However, the monitoring object is not limited to production line 2. For example, the monitoring object may also be a manufacturing device. Figure 1 The abnormality detection device 100 of the embodiment shown calculates a statistic representing the deviation of the actual values ​​of multiple variables relative to the standard values ​​of the multiple variables for each unit time. However, the abnormality detection device 100 may also calculate the statistic for each unit interval. The unit interval is, for example, the period during which the manufacturing device performs the object action.

[0135] (First example of manufacturing device)

[0136] Fig.11 FIG. 1 is a diagram showing an example of a manufacturing device to be monitored. Fig.11 , an injection molding machine 3 is shown as a manufacturing device to be monitored. The injection molding machine 3 includes a molding die 31 , a cylinder 32 , a screw 33 , a hopper 34 , a screw drive device 35 , and a sensor group 36 .

[0137] The cylinder 32 is a cylindrical member having an internal space to which resin is supplied. The hopper 34 supplies resin to the internal space of the cylinder 32. A screw 33 is inserted into the internal space of the cylinder 32. A screw drive device 35 is connected to the base end of the screw 33. The screw 33 is rotated by the control of the screw drive device 35 and can move along the length direction of the cylinder 32. The sensor group 36 measures the values ​​of various parameters indicating the status of the injection molding machine 3.

[0138] The injection molding machine 3 repeatedly performs the injection molding operation. The injection molding operation is performed by moving the screw 33 from Fig.11 The upper state is moved to the lower state, and the resin stored in the internal space of the cylinder 32 is injected into the molding die 31 .

[0139] Fig.12 1 is a graph showing the change in the value of the parameter during the period of performing the injection molding operation. Hereinafter, the period of performing the injection molding operation is referred to as the "injection molding period". Fig.12 In the graph shown, the horizontal axis represents the elapsed time from the start of the injection molding operation, and the vertical axis represents the values ​​of various parameters measured by the sensor group 36. Fig.12As shown, the sensor group 36 measures the screw position, injection pressure and injection speed as parameters representing the status of the injection molding machine 3. The screw position is, for example, a distance D between the front end of the cylinder 32 and the front end of the screw 33 (see Fig.11 ) indicates. Fig.12 In FIG. 3 , line 37 indicates the change of the screw position during injection molding, line 38 indicates the change of the injection pressure, and line 39 indicates the change of the injection speed.

[0140] When the monitoring object is the injection molding machine 3, the abnormality detection device of the second modification obtains standard values ​​of a plurality of variables representing the conditions of the injection molding machine 3 during injection molding when the injection molding machine 3 is operating normally. For example, Figure 4 The simulation execution unit 11 shown uses a virtual model of the injection molding machine 3 to execute a simulation of the values ​​of multiple variables when the injection molding machine 3 is operated according to the production plan, and obtains the values ​​of multiple variables in the injection molding operation obtained by the simulation as standard values. Or, Fig.10 The data set selection unit 17 shown in the figure obtains the values ​​of multiple variables during the injection molding period when the injection molding machine 3 was operating normally in the past as standard values. The multiple variables include, for example, the screw position, injection pressure, and injection speed at the time t1, t2, ..., tn respectively from the start of the injection molding period. In this case, the multiple variables include 3×n variables.

[0141] When the monitoring object is the injection molding machine 3, the collection unit 12 obtains the actual values ​​of multiple variables indicating the status of the injection molding machine 3 during each injection molding period. The collection unit 12 may obtain the actual values ​​of multiple variables from the sensor group 36, or may obtain the actual values ​​of multiple variables from a control device (e.g., PLC) that controls the injection molding machine 3 (not shown).

[0142] Fig.13 It is a graph showing the standard values ​​and actual values ​​of a plurality of variables of the status of the injection molding machine 3. Line 37a shows the standard value of the screw position (n variables) at the time t1, t2, ..., tn from the start of the injection molding period. Line 38a shows the standard value of the injection pressure (n variables) at the time t1, t2, ..., tn. Line 39a shows the standard value of the injection speed (n variables) at the time t1, t2, ..., tn. Line 37b shows the actual value of the screw position (n variables) at the time t1, t2, ..., tn from the start of the injection molding period. Line 38b shows the actual value of the injection pressure (n variables) at the time t1, t2, ..., tn. Line 39b shows the actual value of the injection speed (n variables) at the time t1, t2, ..., tn.

[0143] The calculation unit 13, the detection unit 14, the selection unit 15 and the analysis unit 16 use Fig.13 The standard values ​​and actual values ​​of the 3×n variables shown are processed as described in the above embodiment. Fig.13 As shown, the calculation unit 13 calculates the standard SN ratio as a statistic indicating the degree of deviation of the actual values ​​of the 3×n variables from the standard values ​​of the 3×n variables.

[0144] Fig.14 is a graph showing the change in the statistical quantity calculated for each injection molding operation. Fig.14 , a graph is shown with the number of the injection molding operation as the horizontal axis and the statistic (standard SN ratio) calculated by the calculation unit 13 as the vertical axis. The 7-digit number corresponding to each point in the graph represents the number of the product manufactured by the injection molding operation. Many defective products are confirmed among the products manufactured by the injection molding operation whose standard SN ratio is outside the reference range (i.e., less than the threshold TH). Thus, the user can confirm whether defective products are generated based on the error notification of the detection unit 14.

[0145] Fig.15 This is a graph showing the results of principal component analysis using actual values ​​of effective variables and standard SN ratios for each injection molding period. Fig.15 The biplot of the two principal components PC1 and PC2 selected by the analysis unit 16 is shown in FIG. Fig.15 In the biplot shown, points corresponding to the principal component scores of the two principal components obtained from the actual values ​​of the effective variables of each injection molding action are plotted. Points corresponding to injection molding actions whose standard SN ratio is within the reference range (i.e., above the threshold TH) are concentrated within the frame line 50a. In contrast, points corresponding to injection molding actions whose standard SN ratio is outside the reference range (i.e., less than the threshold TH) are concentrated within the frame line 50b. Thus, the user can confirm the Fig.15 The biplot shown can easily identify abnormal injection molding operation.

[0146] Then, the user confirms Fig.15 The biplot shown in Figure 2 allows us to examine the mechanism of abnormality. Fig.15In the biplot shown, vectors v10, v11, v12, v13, and v14 corresponding to the variables "Speed_time_315 (injection speed at time t315)", "Pressure_time_315 (injection pressure at time t315)", "Speed_time_316 (injection speed at time t316)", "Speed_time_317 (injection speed at time t317)", and "Cylinder_time_318 (screw position at time t318)" respectively face directions different from those corresponding to the other variables. Therefore, the user can understand that the injection speed and screw position near time t315 to t318 contribute greatly to the abnormality of the injection molding machine 3.

[0147] (Second example of manufacturing device)

[0148] Fig.16 FIG. 2 is a diagram showing another example of a manufacturing device to be monitored. Fig.16 , the mounting machine 4 is shown as a manufacturing device to be monitored. The mounting machine 4 includes a chuck 44, a driving device 45, and a sensor group 46.

[0149] The chuck 44 holds the chip 42 to which the bumps 43 are attached. The drive device 45 controls the generation of ultrasonic waves, the lifting and lowering movement of the chuck 44, and the holding force of the chuck 44 on the chip 42. The sensor group 46 measures the values ​​of various parameters indicating the status of the mounting machine 4.

[0150] like Fig.16 As shown, the drive device 45 lowers the chuck 44 holding the chip 42. When the bump 43 contacts the substrate 41, the mounting machine 4 performs a joining action to join the bump 43 to the substrate 41. Specifically, the drive device 45 generates ultrasonic waves to promote the joining of the bump 43 to the substrate 41, and further lowers the chuck 44 while applying a load. When the bump 43 is deformed to a specified amount, the drive device 45 stops the generation of ultrasonic waves and reduces the holding force of the chuck 44 on the chip 42. Thus, the joining action ends. Then, the drive device 45 raises the chuck 44. Each time the chuck 44 holds a new chip 42, the mounting machine 4 performs the above-mentioned joining action.

[0151] The sensor group 46 measures the current and voltage applied to the driving device 45 and the deformation amount of the bump 43 as parameters indicating the condition of the mounting machine 4. The deformation amount of the bump 43 is indicated by the distance between the chip 42 and the substrate 41.

[0152] When the monitoring object is the mounting machine 4, the abnormality detection device of the second modification obtains standard values ​​of a plurality of variables representing the status of the mounting machine 4 during the period of performing the joining operation when the mounting machine 4 is operating normally (hereinafter referred to as "joining operation period"). Figure 4 The simulation execution unit 11 shown uses a virtual model of the mounting machine 4 to execute a simulation of the values ​​of multiple variables when the mounting machine 4 is operated according to the production plan, and obtains the values ​​of multiple variables in the joining operation obtained by the simulation as standard values. Fig.10 The data set selection unit 17 shown in the figure obtains the values ​​of multiple variables during the joining action when the mounting machine 4 was operating normally in the past as standard values. The multiple variables include, for example, the voltage, current, and deformation at each of the elapsed times t1, t2, ..., tm from the start of the joining action. In this case, the multiple variables include 3×m variables.

[0153] When the monitoring object is the mounting machine 4, the collection unit 12 obtains the actual values ​​of multiple variables indicating the status of the mounting machine 4 during each bonding operation. The collection unit 12 may obtain the actual values ​​of multiple variables from the sensor group 46, or may obtain the actual values ​​of multiple variables from a control device (e.g., PLC) that controls the mounting machine 4 (not shown).

[0154] The calculation unit 13, the detection unit 14, the selection unit 15, and the analysis unit 16 perform the processing described in the above embodiment using the standard values ​​and actual values ​​of the 3×m variables. For example, the calculation unit 13 calculates the standard SN ratio as a statistic indicating the degree of deviation of the actual values ​​of the 3×m variables from the standard values ​​of the 3×m variables.

[0155] Fig.17 is a graph showing the change in statistics calculated for each joining action. Fig.17 , a graph is shown with the number of the joining action (installation number) as the horizontal axis and the statistic (standard SN ratio) calculated by the calculation unit 13 as the vertical axis. Many defective products are confirmed in the products manufactured by the joining action whose standard SN ratio is outside the reference range (i.e., less than the threshold TH). Thus, the user can confirm whether defective products are generated based on the error notification of the detection unit 14.

[0156] (Another example of a unit section)

[0157] In the above example, the unit interval is set as the period during which the action of the manufacturing device (injection molding action or bonding action) is performed. However, the unit interval is not limited to this. For example, the unit interval may also be each of a plurality of intervals included in the range of the physical quantity of the manufacturing device.

[0158] For example, in manufacturing equipment Fig.11In the case of the injection molding machine 3 shown, the screw position can be varied from Fig.11 The position of the screw 33 shown in the upper part (indicated by the distance D1) is Fig.11 The range of the position of the screw 33 shown in the lower part (indicated by the distance D2). In this case, a plurality of intervals included in the range of the distance D1 to the distance D2 are respectively set as unit intervals. For example, the intervals of the distance D from D1 to D3 and the intervals of the distance D from D3 to D2 are set as unit intervals.

[0159] Alternatively, the interval in which the physical quantity obtained from the tensile tester or compression tester installed at the production site falls within a specific range may be set as the unit interval. For example, the interval in which the strain observed in the stress-strain diagram falls within a specified range may be set as the unit interval. Alternatively, in the relationship between the cooling rate and hardness of the metal heated at the production site, the interval in which the cooling rate falls within a specified range may be set as the unit interval.

[0160] §3 Notes

[0161] As described above, this embodiment includes the following disclosure.

[0162] (Structure 1)

[0163] The abnormality detection device (100, 100A) comprises:

[0164] An acquisition unit (101, 11, 17) that acquires standard values ​​per unit time of a plurality of variables representing the status of the production line (2) when the production line (2) is operating normally;

[0165] A collecting unit (101, 12) for collecting actual values ​​of the plurality of variables at each unit time;

[0166] a calculation unit (101, 13) for calculating, for each unit time, a statistic representing the degree of deviation of the actual values ​​of the plurality of variables from the standard values ​​of the plurality of variables; and

[0167] A detection unit (101, 14) detects that an abnormality has occurred in the production line based on the fact that the statistic is outside a reference range.

[0168] (Structure 2)

[0169] The abnormality detection device (100, 100A) comprises:

[0170] An acquisition unit (101, 11, 17) that acquires standard values ​​of a plurality of variables representing the conditions of the monitoring objects (2, 3, 4) in a unit section when the monitoring objects (2, 3, 4) in a production site are operating normally;

[0171] A collecting unit (101, 12) for collecting actual values ​​of the plurality of variables in the unit interval;

[0172] a calculation unit (101, 13) for calculating a statistic representing a degree of deviation of actual values ​​of the plurality of variables from the standard values ​​of the plurality of variables; and

[0173] A detection unit (101, 14) detects that an abnormality has occurred in the monitored object based on the fact that the statistic is outside a reference range.

[0174] (Structure 3)

[0175] In the abnormality detection device described in structure 1 or 2,

[0176] The abnormality detection device further includes a selection unit (101, 15) for selecting an effective variable having a relatively large influence on the statistic from among the plurality of variables.

[0177] (Structure 4)

[0178] In the abnormality detection device (100, 100A) described in Configuration 3,

[0179] The selection unit (101, 15) selects, as the effective variable, a variable that is effective in improving the estimation accuracy of a T-method estimation model that uses the plurality of variables as explanatory variables and the statistic as a target variable.

[0180] (Structure 5)

[0181] In the abnormality detection device (100, 100A) described in Structure 3 or 4,

[0182] The abnormality detection device (100, 100A) further includes an analysis unit (101, 16) that performs a principal component analysis using the actual value of the effective variable for each unit time and the statistic.

[0183] (Structure 6)

[0184] In the abnormality detection device (100, 100A) described in Structure 3 or 4,

[0185] The abnormality detection device (100, 100A) further includes an analysis unit (101, 16) that performs a principal component analysis using the actual value of the effective variable and the statistic for each of the unit intervals.

[0186] (Structure 7)

[0187] In the abnormality detection device (100, 100A) described in Structure 5 or 6,

[0188] The analysis unit (101, 16) selects two principal components that maximize the variance of the statistic,

[0189] The analyzing unit (101, 16) determines the influence of the effective variable on the standard SN ratio based on the principal component loadings of the effective variable and the statistic with respect to the two principal components.

[0190] (Structure 8)

[0191] In the abnormality detection device (100) described in any one of the configurations 1 and 3 to 7,

[0192] The acquisition unit (101, 11) uses a virtual model (2M) of the production line (2) to perform a simulation of the values ​​of the plurality of variables when the production line (2) is operated according to a production plan.

[0193] The acquisition unit (101, 11) acquires the value for each unit time obtained by the simulation as the standard value.

[0194] (Structure 9)

[0195] In the abnormality detection device (100) described in any one of Configurations 2 to 7,

[0196] The acquisition unit (101, 11) uses a virtual model (2M) of the monitored object (2, 3, 4) to perform a simulation of the values ​​of the plurality of variables when the monitored object (2, 3, 4) is operated according to a production plan.

[0197] The acquisition unit (101, 11) acquires the value of the unit section obtained by the simulation as the standard value.

[0198] (Structure 10)

[0199] In the abnormality detection device (100A) described in any one of Configurations 1 and 3 to 7,

[0200] The acquisition unit (101, 17) acquires, as the standard value, the values ​​per unit time of the plurality of variables when the production line (2) was operating normally in the past.

[0201] (Structure 11)

[0202] In the abnormality detection device (100A) described in any one of Configurations 2 to 7,

[0203] The acquisition unit (101, 17) acquires, as the standard values, the values ​​of the plurality of variables in the unit section when the monitored object (2, 3, 4) was operating normally in the past.

[0204] (Structure 12)

[0205] In the abnormality detection device (100, 100A) described in any one of the structures 1 to 11,

[0206] The statistic is the standard SN ratio.

[0207] (Structure 13)

[0208] In the abnormality detection device (100, 100A) described in any one of the structures 1 and 3 to 12,

[0209] The monitoring object is a production line (2),

[0210] The unit interval is a unit time.

[0211] (Structure 14)

[0212] In the abnormality detection device (100, 100A) described in any one of Configurations 2 to 12,

[0213] The monitoring object includes manufacturing equipment (3, 4).

[0214] (Structure 15)

[0215] In the abnormality detection device (100, 100A) described in configuration 14,

[0216] The manufacturing device (3, 4) repeatedly performs the target action,

[0217] The unit section is a period during which the target action is performed.

[0218] (Structure 16)

[0219] The anomaly detection method has the following steps:

[0220] Step (S1), obtaining standard values ​​per unit time of a plurality of variables representing the status of the production line (2) when the production line (2) is operating normally;

[0221] Step (S2), collecting actual values ​​of the plurality of variables at each unit time;

[0222] Step (S3), calculating, for each of the unit time, a statistic representing the degree of deviation of the actual values ​​of the plurality of variables relative to the standard values ​​of the plurality of variables; and

[0223] Step (S6), detecting that an abnormality has occurred in the production line based on the fact that the statistic is outside a reference range.

[0224] (Structure 17)

[0225] A program (131, 132) causes a computer (101) to execute an abnormality detection method, wherein:

[0226] The anomaly detection method comprises the following steps:

[0227] Step (S1), obtaining standard values ​​per unit time of a plurality of variables representing the status of the production line (2) when the production line (2) is operating normally;

[0228] Step (S2), collecting actual values ​​of the plurality of variables at each unit time;

[0229] Step (S3), calculating, for each of the unit time, a statistic representing the degree of deviation of the actual values ​​of the plurality of variables relative to the standard values ​​of the plurality of variables; and

[0230] Step (S6), detecting that an abnormality has occurred in the production line based on the fact that the statistic is outside a reference range.

[0231] (Structure 18)

[0232] The anomaly detection method has the following steps:

[0233] Step (S1), obtaining standard values ​​of a plurality of variables representing the conditions of the monitoring objects (2, 3, 4) in a unit section when the monitoring objects (2, 3, 4) in the production site are operating normally;

[0234] Step (S2), collecting actual values ​​of the plurality of variables in the unit interval;

[0235] Step (S3), calculating a statistic representing the deviation of the actual values ​​of the plurality of variables relative to the standard values ​​of the plurality of variables; and

[0236] Step (S6) of detecting that an abnormality has occurred in the monitored object based on the fact that the statistic is outside a reference range.

[0237] (Structure 19)

[0238] A program (131, 132) causes a computer to execute an abnormality detection method, wherein:

[0239] The anomaly detection method comprises the following steps:

[0240] Step (S1), obtaining standard values ​​of a plurality of variables representing the conditions of the monitoring objects (2, 3, 4) in a unit section when the monitoring objects (2, 3, 4) in the production site are operating normally;

[0241] Step (S2), collecting actual values ​​of the plurality of variables in the unit interval;

[0242] Step (S3), calculating a statistic representing the deviation of the actual values ​​of the plurality of variables relative to the standard values ​​of the plurality of variables; and

[0243] Step (S6) of detecting that an abnormality has occurred in the monitored object based on the fact that the statistic is outside a reference range.

[0244] Although the embodiments of the present invention have been described, the embodiments disclosed this time should be considered to be illustrative in all aspects and not restrictive. The scope of the present invention is indicated by the claims, and it is intended to include all modifications within the meaning and scope equivalent to the claims.

[0245] Description of symbols

[0246] 1: System; 2: Production line; 2M: Virtual model; 3: Injection molding machine; 4: Mounting machine; 11: Simulation execution unit; 12: Collection unit; 13: Calculation unit; 14: Detection unit; 15: Selection unit; 16: Analysis unit; 17: Data set selection unit; 18: Live database; 21: Cutting process; 21M: First model; 21a: Cutting device; 21b, 22c: Timer; 21c, 21d, 22d, 23c: Counter; 22: Parts delivery process; 22M: Second model; 22a, 23a, 23b: Sensor; 22b: Automatic guided vehicle; 23: Assembly process; 23 M: the third model; 24: product shipping process; 30: chart; 31: molding die; 32: cylinder; 33: screw; 34: hopper; 35: screw drive device; 36, 46: sensor group; 40: double plot; 41: substrate; 42: chip; 43: bump; 44: chuck; 45: drive device; 70: display device; 75: input device; 100, 100A: abnormality detection device; 101: processor; 102: memory; 103: storage; 104: display controller; 105: input interface; 106: communication interface; 131: abnormality detection program; 132: simulation program.

Claims

1. An abnormality detection device, wherein: The abnormality detection device has: an acquisition unit that acquires standard values ​​of a plurality of variables indicating the status of the monitored object in a unit section when the monitored object in the production site operates normally; A collecting unit, which collects actual values ​​of the plurality of variables in the unit interval; a calculation unit that calculates a statistic indicating a degree of deviation of actual values ​​of the plurality of variables from the standard values ​​of the plurality of variables; as well as The detection unit detects that an abnormality has occurred in the monitored object based on the fact that the statistic is outside a reference range.

2. The abnormality detection device according to claim 1, wherein: The abnormality detection device further includes a selection unit that selects a valid variable having a relatively large influence on the statistic from among the plurality of variables.

3. The abnormality detection device according to claim 2, wherein: The selection unit selects, as the effective variable, a variable effective for improving the estimation accuracy of a T-method estimation model that uses the plurality of variables as explanatory variables and the statistic as a target variable.

4. The abnormality detection device according to claim 2, wherein: The abnormality detection device further includes an analysis unit that performs a principal component analysis using the actual value of the effective variable and the statistic for each unit interval.

5. The abnormality detection device according to claim 4, wherein: The analyzing unit selects two principal components that maximize the variance of the statistic, The analyzing unit determines the degree of influence of the effective variable on the statistic based on the principal component loadings of each of the effective variable and the statistic with respect to the two principal components.

6. The abnormality detection device according to claim 1, wherein: The acquisition unit performs a simulation of the values ​​of the plurality of variables when the monitored object is operated according to a production plan using a virtual model of the monitored object. The acquisition unit acquires the value of the unit section obtained by the simulation as the standard value.

7. The abnormality detection device according to claim 1, wherein: The acquisition unit acquires, as the standard value, values ​​of the plurality of variables in the unit section when the monitored object was operating normally in the past.

8. The abnormality detection device according to claim 1, wherein: The statistic is the standard SN ratio.

9. The abnormality detection device according to any one of claims 1 to 8, wherein: The monitoring object is a production line, The unit interval is a unit time.

10. The abnormality detection device according to any one of claims 1 to 8, wherein: The monitoring object includes a manufacturing device.

11. The abnormality detection device according to claim 10, wherein: The manufacturing device repeatedly performs the target action, The unit section is a period during which the target action is performed.

12. An anomaly detection method comprising the following steps: acquiring standard values ​​of a plurality of variables representing the status of the monitored object in a unit section when the monitored object in the production site operates normally; collecting actual values ​​of the plurality of variables in the unit interval; Calculating a statistic representing the deviation of the actual values ​​of the plurality of variables from the standard values ​​of the plurality of variables; as well as When the statistic is outside a reference range, it is detected that an abnormality has occurred in the monitored object.

13. A program causing a computer to execute an abnormality detection method, wherein: The anomaly detection method comprises the following steps: acquiring standard values ​​of a plurality of variables representing the status of the monitored object in a unit section when the monitored object in the production site operates normally; collecting actual values ​​of the plurality of variables in the unit interval; Calculating a statistic representing the deviation of the actual values ​​of the plurality of variables from the standard values ​​of the plurality of variables; as well as When the statistic is outside a reference range, it is detected that an abnormality has occurred in the monitored object.

Images