Production support system, production support method, and program

Abstract

A production support system (1), which estimates the probability that each of a plurality of units becomes a cause of an installation error for an installation error occurring in an installation machine (10) comprising a plurality of units, is provided with: an acquisition unit (for example, an error count pre-processing unit (31) and a correction amount pre-processing unit (32)) that acquires an installation log for the object of the estimated probability from the installation machine (10), and acquires a correction amount of the installation log for the object of the estimated probability from the installation machine (10); the installation log comprises information related to an installation error; an installation error cause estimation unit (33) that estimates the probability of each of the plurality of units on the basis of the installation log and a first estimation model that is a model that is based on the relationship between the number of installation errors of each of the plurality of units and the abnormality of the unit; and an output unit (for example, a display device (40)) that outputs the estimation result of the installation error cause estimation unit (33). The installation log includes information relating to the number of production in the plurality of units and the number of installation errors.

Description

Technical Field

[0001] The present disclosure relates to a production assistance system, a production assistance method, and a program. Background Art

[0002] In Patent Document 1, a mounting substrate manufacturing system is disclosed, which can effectively utilize tracking information, predict changes in manufacturing quality and equipment status, and appropriately set parameters of mounting equipment (mounter), thereby obtaining good manufacturing quality.

[0003] Prior Art Documents

[0004] Patent Documents

[0005] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2021-12979 Summary of the Invention

[0006] Problems to be Solved by the Invention

[0007] However, there are times when it is desired to assist in determining the units that are the causes of mounting errors in the mounter while suppressing cost increases.

[0008] To this end, the present disclosure provides a production assistance system, a production assistance method, and a program that can assist in determining the units that are the causes of mounting errors while suppressing cost increases.

[0009] Means for Solving the Problems

[0010] The production assistance system according to one aspect of the present disclosure estimates the probability that each of a plurality of units becomes a cause of a mounting error occurring in a mounter composed of the plurality of units. The production assistance system includes: an acquisition unit that acquires a first mounting log of an object for estimating the probability from the mounter, the first mounting log including information related to the mounting error; a mounting error cause estimation unit that estimates the probability of each of the plurality of units based on the first mounting log and a first estimation model, the first estimation model being a model based on the relationship between the number of mounting errors of each of the plurality of units and the abnormality of the unit; and an output unit that outputs the estimation result of the mounting error cause estimation unit. The first mounting log includes information related to the production quantity and the number of mounting errors among the plurality of units.

[0011] A production assistance method according to an aspect of the present disclosure estimates the probability that each of a plurality of units is a cause of an installation error occurring in an installation machine composed of the plurality of units. In the production assistance method, a first installation log for estimating the probability is obtained from the installation machine, the first installation log including information related to the installation error, and the probability of each of the plurality of units is estimated based on the first installation log and a first estimation model, the first estimation model being a model based on the relationship between the number of installation errors of each of the plurality of units and the abnormality of the unit, and the estimated estimation result is output. The first installation log includes information related to the production quantity and the number of installation errors among the plurality of units.

[0012] A program according to an aspect of the present disclosure is a program for causing a computer to execute the above-described production assistance method.

[0013] Advantageous Effects of the Invention

[0014] According to an aspect of the present disclosure, it is possible to realize a production assistance system or the like that can assist in determining a unit that is a cause of an installation error while suppressing an increase in cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a block diagram showing a functional configuration of a production assistance system according to an embodiment.

[0016] Figure 2 is a diagram showing a first example of data after preprocessing according to an embodiment.

[0017] Figure 3 is a diagram showing a second example of data after preprocessing according to an embodiment.

[0018] Figure 4 is a diagram showing an example of first regression coefficient information according to an embodiment.

[0019] Figure 5 is a diagram showing an example of a correspondence table between tape feed accuracy and accuracy grade according to an embodiment.

[0020] Figure 6 is a first flowchart showing the operation of a production assistance system according to an embodiment.

[0021] Figure 7 is a diagram for explaining the estimation and use of a first regression coefficient according to an embodiment.

[0022] Figure 8 is a diagram for explaining the estimation and use of a second regression coefficient according to an embodiment.

[0023] Figure 9 It is the second flowchart showing the operation of the production support system according to the embodiment.

[0024] Figure 10 It is a diagram showing a first example of a screen displayed on the display device according to the embodiment.

[0025] Figure 11 It is a diagram showing a second example of a screen displayed on the display device according to the embodiment. Detailed Embodiment

[0026] A production support system according to an aspect of the present disclosure estimates the probability that each of a plurality of units is a cause of an installation error occurring in an installation machine composed of the plurality of units. The production support system includes: an acquisition unit that acquires a first installation log of an object for estimating the probability from the installation machine, the first installation log including information related to the installation error; an installation error cause estimation unit that estimates the probability of each of the plurality of units based on the first installation log and a first estimation model, the first estimation model being a model based on the relationship between the number of installation errors of each of the plurality of units and the abnormality of the unit; and an output unit that outputs the estimation result of the installation error cause estimation unit. The first installation log includes information related to the production quantity and the number of installation errors among the plurality of units.

[0027] Accordingly, since the estimated probabilities of the plurality of units are output as the estimation result, a manager or the like of the installation machine can determine an abnormal unit with reference to the estimation result. That is, the production support system can assist a manager or the like in determining an abnormal unit by outputting the estimation result. In addition, since information related to the production quantity and the number of installation errors output as a standard function of the installation machine is used, it is possible to estimate the probability without adding sensors or the like to the installation machine. Therefore, the production support system according to an aspect of the present disclosure can assist in determining a unit that is a cause of an installation error while suppressing an increase in cost.

[0028] In addition, for example, the first estimation model may be a multiple regression model, and the multiple regression model uses the first regression coefficients of the plurality of units based on the distribution of the information related to the production quantity and the number of installation errors of the plurality of units as explanatory variables and the probability as the target variable.

[0029] Accordingly, since the probability is estimated considering multiple factors (multiple units), the estimation accuracy of the probability can be improved.

[0030] Alternatively, for example, it may also be that the production assistance system further includes: a first coefficient estimation unit that estimates the first regression coefficient of each of the plurality of units based on information related to the production quantity and the distribution of the number of installation errors included in the second installation log of the installation machine obtained before the first installation log.

[0031] Thus, the production assistance system can perform a series of processes from the generation of the first estimation model to the estimation of the probability.

[0032] Alternatively, for example, it may also be that the output unit displays the probability of each of the plurality of units for the installation error.

[0033] Thus, since the cause probability of the installation error can be visualized, when determining the abnormality of multiple units, the manager of the installation machine, etc. can refer to the displayed information. Therefore, it is possible to effectively assist in determining the unit that is the cause of the installation error.

[0034] Alternatively, for example, it may also be that the plurality of units include a belt feeder for supplying components and a nozzle for adsorbing the components, and the production assistance system further includes: a belt feed accuracy estimation unit that estimates the belt feed accuracy based on the first installation log and a second estimation model, where the second estimation model is a model based on the relationship between the belt feed accuracy of the belt feeder and the amount related to the adsorption position deviation of the component, and the first installation log further includes information related to the control amount for controlling the nozzle.

[0035] Thus, the manager of the installation machine, etc. can further refer to the information based on the estimated belt feed accuracy to determine the abnormal unit. That is, by outputting the information based on the belt feed accuracy, the production assistance system can further assist the manager, etc. in determining the abnormal unit while suppressing the increase in the processing amount related to the estimation of the belt feed accuracy.

[0036] Alternatively, for example, it may also be that the second estimation model is a simple regression model, and the simple regression model includes a second regression coefficient representing the relationship between the belt feed accuracy and the amount related to the adsorption position deviation of the component.

[0037] Thus, the production assistance system can further assist the manager, etc. in determining the abnormal unit while suppressing the increase in the processing amount related to the estimation of the belt feed accuracy.

[0038] Alternatively, for example, it may also be that the production assistance system further includes: a second coefficient estimation unit that estimates the second regression coefficient of each of the plurality of units based on the control amount for controlling the nozzle and the distribution of the measured values of the belt feed accuracy of the belt feeder included in the third installation log of the installation machine obtained before the first installation log.

[0039] Accordingly, the production assistance system can execute a series of processes from the generation of the second estimation model to the estimation of the tape feed accuracy.

[0040] In addition, for example, the output unit may display information related to the tape feed accuracy estimated by the tape feed accuracy estimation unit.

[0041] Accordingly, since the information related to the tape feed accuracy can be visualized, managers of the mounter or the like can refer to the displayed information when determining the abnormality of the feeder. Therefore, it is possible to effectively assist in determining the units that are the causes of mounting errors.

[0042] In addition, for example, the mounter may not be equipped with sensors for directly measuring the states of the plurality of units.

[0043] Accordingly, since the mounter may not be equipped with hard sensors, it is possible to more reliably suppress the cost of the mounter.

[0044] In addition, for example, the plurality of units may include a head spindle, a nozzle, and a feeder, and the mounting error cause estimation unit estimates the probabilities of the head spindle, the nozzle, the feeder, and the components supplied by the feeder.

[0045] Accordingly, the production assistance system can estimate the probabilities of the plurality of units including the head spindle, the nozzle, and the feeder and the components. In other words, the production assistance system can assist in the judgment by managers or the like as to which of the head spindle, the nozzle, the feeder, and the components is the cause of the mounting error.

[0046] In addition, the production assistance method according to one aspect of the present disclosure estimates the probability that each of the plurality of units is the cause of a mounting error occurring in a mounter composed of a plurality of units. In the production assistance method, a first mounting log for estimating the probability is obtained from the mounter, the first mounting log includes information related to the mounting error, the probabilities of the plurality of units are estimated based on the first mounting log and a first estimation model, the first estimation model is a model based on the relationship between the number of mounting errors of each of the plurality of units and the abnormality of the unit, the estimated estimation result is output, and the first mounting log includes information related to the production quantity and the number of mounting errors among the plurality of units. In addition, the program according to one aspect of the present disclosure is a program for causing a computer to execute the above-described production assistance method.

[0047] Accordingly, the same effects as those of the above-described production assistance system are achieved.

[0048] In addition, these holistic or specific manners can be implemented by a system, a method, an integrated circuit, a computer program, or a non-transitory recording medium such as a computer-readable CD-ROM, or can be implemented by any combination of a system, a method, an integrated circuit, a computer program, or a recording medium. The program can be pre-stored in the recording medium or can be supplied to the recording medium via a wide-area communication network including the Internet and the like.

[0049] Hereinafter, embodiments will be specifically described with reference to the drawings.

[0050] In addition, the embodiments described below all represent general or specific examples. The numerical values, shapes, structural elements, arrangement positions of the structural elements, connection manners, steps, order of steps, etc. shown in the following embodiments are examples and are not intended to limit the present disclosure. In addition, among the structural elements in the following embodiments, the structural elements not described in the independent claims are described as optional structural elements.

[0051] In addition, the drawings are schematic diagrams and are not necessarily drawn precisely. Therefore, for example, the scales in the drawings are not necessarily the same. In addition, in the drawings, the same reference numerals are given to substantially the same structures, and repeated descriptions are omitted or simplified.

[0052] In addition, in this specification, terms indicating the relationship between elements being the same or the like, as well as numerical values and numerical ranges, are not expressions indicating only a strict meaning, but mean expressions including substantially the same ranges, for example, a difference of about several % (or, about 10%).

[0053] In addition, in this specification, ordinal numbers such as "first", "second", etc. do not mean the number or order of structural elements unless otherwise specified, but are used for the purpose of distinguishing the same kind of structural elements to avoid confusion.

[0054] (Embodiment)

[0055] Hereinafter, with reference to Figures 1 to 11 a production assistance system including a production assistance device according to the present embodiment will be described.

[0056] [1. Structure of Production Assistance System]

[0057] First, with reference to Figure 1 the structure of the production assistance system according to the present embodiment will be described. Figure 1 is a block diagram showing the functional structure of the production assistance system 1 according to the present embodiment.

[0058] As shown in Figure 1As shown, the production assistance system 1 includes a coefficient estimation device 20, a cause probability estimation device 30, and a display device 40. The production assistance system 1 is a system for assisting production using the mounter 10. Specifically, the production assistance system 1 indirectly estimates the cause probability (also referred to as the error cause probability or the mounting error cause probability) of the unit (equipment unit) for the mounting error generated in the mounter 10 without using the data of sensors that directly sense the unit. In addition, in the present embodiment, the production assistance system 1 further indirectly estimates the tape feeding accuracy of the feeder without using the data of sensors that directly sense the tape feeding accuracy of the feeder. The sensors are, for example, sensors that measure multiple units, such as sensors that measure the supply position of components of the feeder, the mounting position of components, etc.

[0059] The mounter 10 is an example of production equipment that constitutes a manufacturing production line and is a mounting device (component mounting device) for mounting components on an object (workpiece) such as a substrate. The mounter 10 is composed of multiple units. In the present embodiment, the mounter 10 has a drive control unit (hereinafter also referred to as the main shaft), a component assembly head unit (hereinafter also referred to as the nozzle), and a component supply unit (hereinafter also referred to as the feeder) as units. The drive control unit controls the rotation and movement of the component assembly head unit. The drive control unit has a head main shaft that controls the rotation of the component assembly head unit by driving a motor. The component assembly head unit assembles (mounts) the components supplied to the component supply position (component adsorption position) of the feeder on the substrate. The component assembly head unit is configured to include an assembly head (head), etc., and the assembly head is equipped with component adsorption nozzles (nozzles) that can adsorb components from the feeder and lift and lower individually. For example, multiple nozzles are assembled in the head. The component supply unit arranges and configures one or more feeders and supplies components to the component supply position respectively. In the present embodiment, the feeder is a tape feeder.

[0060] The mounter 10 outputs the first mounting log L1 to the cause probability estimation device 30 regularly during production. The mounter 10 can output the first mounting log L1, for example, every time one component is mounted on the substrate, or every time a given number of components are mounted on the substrate.

[0061] In addition, the first installation log L1 includes information related to the control of each unit and information indicating production performance obtained during the period from the operation of sucking the component from the self-feeder to the operation of mounting the component on the substrate in the mounter 10. The first installation log L1 is a log output by the mounter 10 as a standard function. For example, in the first installation log L1, it includes information related to the control amounts for controlling multiple units, information identifying the multiple units used in production, and information related to production performance. Among the information related to the control amounts, for example, it includes information related to the control amounts (correction amounts) for controlling the movement and rotation of a unit (e.g., a nozzle) (such as the adsorption correction amount and the recognition correction amount described later). Among the information indicating production performance, it includes information related to the production quantity (e.g., the number of adsorption trials) and the number of mounting errors. In addition, in the first installation log L1, it does not include data obtained by directly sensing the unit (i.e., the data of the hard sensor).

[0062] In addition, the number of mounters 10 assisted by the production support system 1 is not particularly limited and may be one or multiple.

[0063] In addition, the component is an electronic component, such as a resistor, a capacitor, etc., but is not limited thereto. In addition, the object is not limited to the substrate and may be any workpiece that can be subjected to a given process.

[0064] In addition, the mounter 10 may not have a sensor (hard sensor) that directly measures the states (e.g., operations) of the drive control unit, the component assembly head unit, and the component supply unit. Such a sensor includes, for example, a sensor that is not standardly equipped on the mounter 10 and is set through subsequent installation. Whether it is standardly equipped can be confirmed through the catalog of the mounter 10, etc.

[0065] The coefficient estimation device 20 uses the second installation log L2 obtained in advance from the mounter 10 to perform a process of estimating each regression coefficient. Here, "in advance" means before performing the above-mentioned estimation process (for example, before obtaining the first installation log L1). In addition, the second installation log L2 is an installation log for pre-estimating each regression coefficient. Furthermore, the second installation log L2 is the same data as the first installation log L1, and includes information related to the control of each unit obtained during the period from the operation of sucking a component from the feeder to the operation of mounting the component on the substrate in the mounter 10. The second installation log L2 is a log output by the mounter 10 as a standard function. For example, in the second installation log L2, information related to the control amount for controlling a plurality of units, the number of installation errors, information for identifying a plurality of units used in production, and information related to production results are included. In the information related to the control amount, for example, information related to the correction amount for correcting the movement and rotation of a unit (for example, a nozzle) (for example, the adsorption correction amount and the recognition correction amount described later) is included. In the information related to the production results, the production quantity (for example, the number of adsorption trials) and the number of installation errors are included. In addition, in the second installation log L2, data obtained by directly sensing a unit (that is, data of a hard sensor) is not included. The second installation log L2 can also be used as the third installation log.

[0066] The coefficient estimation device 20 includes: an error count preprocessing unit 21, a correction amount preprocessing unit 22, a non-linear regression coefficient estimation unit 23, and a linear regression coefficient estimation unit 24. The coefficient estimation device 20 can be implemented by a CPU (Central Processing Unit) and a memory, etc. In addition, the processing of each functional block of the coefficient estimation device 20 is generally realized by a program execution unit such as a processor reading and executing software (program) recorded in a recording medium such as a ROM.

[0067] The error count preprocessing unit 21 obtains the second installation log L2 from the mounter 10, extracts data for estimating the first regression coefficient in the non-linear regression coefficient estimation unit 23 from the second installation log L2, and outputs the extracted preprocessed data to the non-linear regression coefficient estimation unit 23. Figure 2 It is a diagram showing a first example of the preprocessed data according to the present embodiment. Figure 2 It shows an example of the preprocessed data output from the error count preprocessing unit 21 to the non-linear regression coefficient estimation unit 23.

[0068] As Figure 2As shown, the pre-processed data includes "index", "feeder serial number", "head spindle number", "nozzle serial number", "component serial number", "component size", "number of adsorption trials", and "number of installation errors".

[0069] The index is a number assigned for each combination of "feeder", "head spindle", "nozzle", and "component size". That is, the pre-processed data is data aggregated for each combination of "feeder", "head spindle", "nozzle", and "component size".

[0070] The feeder serial number is the identification information (e.g., identification number) of each feeder.

[0071] The head spindle number is the identification number for identifying the head spindle that controls the rotation of the nozzle.

[0072] The nozzle serial number is the identification information (e.g., identification number) of each nozzle. In one component mounting head unit, multiple nozzles are assembled, and different identification information is assigned to each of the multiple nozzles.

[0073] The component size represents the width of the component used in production. Figure 2 The "1.0" shown, for example, means that the width is 1 mm when observing the component from the upper surface. In addition, the width of the component refers to the length in the direction (X direction) orthogonal to the traveling direction (Y direction) of the component in the feeder. The X direction is the conveyance direction of the substrate in the mounter 10.

[0074] The component serial number is the identification information (e.g., identification number) for identifying the component supplied by the feeder.

[0075] The number of adsorption trials represents the number of times the nozzle has adsorbed the component. The number of adsorption trials is the number of times the nozzle performs component adsorption during production in a given period or a given number (e.g., 1 lot).

[0076] The number of installation errors represents the number of installation errors that occurred among the number of adsorption trials. Installation errors include adsorption errors of the nozzle to the component, installation errors of the component relative to the substrate, etc.

[0077] For example, index "1" means that when the nozzle with the identification information "NZ00001" controlled by the head spindle with the identification number "1" adsorbs the component with the identification information "CP0001" and size "1.0" 1000 times from the feeder with the identification information "FD00001", 12 installation errors occurred. Thus, in this pre-processed data, information identifying the unit where the component was installed, information identifying the component to be installed, and information related to production results are included.

[0078] Details will be described later. However, the inventors of the present application found that there is a relationship between the number of installation errors and the abnormality of the unit. That is, the inventors of the present application found that the abnormality of the unit can be estimated based on the number of installation errors. Therefore, in the pre-processed data output from the error count pre-processing unit 21, the number of installation errors is essential information. In addition, the so-called abnormality refers to an alarm that may not be accompanied by the stop of the installation machine 10. For example, it is a state where it is presumed that a failure or deterioration has occurred in at least one of the feeder, the head spindle, and the nozzle.

[0079] Refer again to Figure 1 , the correction amount pre-processing unit 22 obtains the second installation log L2 from the installation machine 10, extracts data for estimating the second regression coefficient in the linear regression coefficient estimation unit 24 from the second installation log L2, and outputs the extracted pre-processed data to the linear regression coefficient estimation unit 24. Figure 3 FIG. is a diagram showing a second example of the pre-processed data according to the present embodiment. Figure 3 Shows an example of the pre-processed data output from the correction amount pre-processing unit 22 to the linear regression coefficient estimation unit 24.

[0080] As Figure 3 shown, in the pre-processed data, it includes "index", "feeder serial number", "head spindle number", "nozzle serial number", "component serial number", "component size", "adsorption trial number", and "each correction amount".

[0081] Each correction amount includes an identification correction amount and an adsorption correction amount. In the identification correction amount and the adsorption correction amount, it includes a correction amount related to the X coordinate, a correction amount related to the Y coordinate, and the average value, median value, and standard deviation of the correction amount. The identification correction amount represents the correction amount of the position of the nozzle in the X-axis direction and the Y-axis direction after the nozzle adsorbs the component. The identification correction amount represents, for example, the adjustment amount of the position of the nozzle in the X-axis direction and the Y-axis direction for mounting the component on the mounting target position of the substrate. For example, the identification correction amount is a value based on the difference between the position where the nozzle adsorbs the component and the adsorption target position.

[0082] The adsorption correction amount represents the correction amount of the position of the nozzle in the X-axis direction and the Y-axis direction when the nozzle adsorbs the component. The adsorption correction amount represents, for example, the adjustment amount when the position of the nozzle is adjusted to the adsorption target position (for example, the position at the center of the upper surface of the component) where the nozzle can adsorb the component. For example, the adsorption correction amount is a value based on the difference between the adsorption target position of the component supplied by the feeder and the adsorption center of the nozzle.

[0083] In addition, in each of the recognition correction amount and the adsorption correction amount, it is sufficient that at least one of the average value, the median value, and the standard deviation is included in the data after preprocessing. Further, each correction amount can be determined from an image captured by the imaging device provided in the mounting machine 10.

[0084] In addition, the correction amount preprocessing unit 22 may also calculate an adsorption position deviation statistic described later based on the recognition correction amount and the adsorption correction amount. The adsorption position deviation statistic may also be included in the data after preprocessing.

[0085] Details will be described later, but the inventors of the present application found that there is a relationship between each correction amount and the tape feeding accuracy. That is, the inventors of the present application found that the tape feeding accuracy can be estimated based on each correction amount. Therefore, in the data after preprocessing output from the correction amount preprocessing unit 22, each correction amount is essential information.

[0086] Refer again to Figure 1 , the non-linear regression coefficient estimation unit 23 estimates a first regression coefficient for estimating the cause probability of a unit with a mounting error based on the data after preprocessing shown in Figure 2 . As main elements in the component mounting process, there are a feeder, a head spindle, a nozzle, and a component size. The first regression coefficient is a coefficient for identifying which of the elements of the feeder, the head spindle, the nozzle, and the component size causes the cause of the mounting error included in the first mounting log L1. The first mounting log L1 is a log including information of the same items as those in the second mounting log L2. The non-linear regression coefficient estimation unit 23 is an example of the first coefficient estimation unit.

[0087] In the present embodiment, as a method for obtaining the respective cause probabilities in the case where a mounting error occurs based on the feeder, the head spindle, the nozzle, and the component size involved in the component mounting process, the mounting error cause estimation unit 33 of the cause probability estimation device 30 uses "logistic regression analysis" which is one of non-linear regression analyses. Therefore, the non-linear regression coefficient estimation unit 23 estimates the first regression coefficient used in the logistic regression analysis. Specifically, the non-linear regression coefficient estimation unit 23 estimates the first regression coefficient of the linear predictor included in the logistic function (refer to Equation 2 described later) used in the logistic regression analysis.

[0088] The non-linear regression coefficient estimation unit 23 estimates the first regression coefficient of each element based on the distribution of the number of adsorption trials and the number of installation errors for combinations of elements such as the feeder, the head spindle, the nozzle, and the component size, using the Markov Chain Monte Carlo (MCMC) method. The MCMC method is a technique for obtaining samples (generating random numbers) from a multivariate probability distribution and is a technique often used when performing maximum likelihood estimation of multiple parameters (here, partial regression coefficients) that make up a statistical model. Each element includes, for example, the feeder, the head spindle, the nozzle, and the component (component size). In addition, the first regression coefficient is an explanatory variable in the first estimation model 33a.

[0089] Figure 4 It is a diagram showing an example of the first regression coefficient information related to the present embodiment.

[0090] As Figure 4 shown, it includes "index", "feeder serial number", "head spindle number", "nozzle serial number", "component serial number", "component size", "number of adsorption trials", "number of installation errors", "bias term", and "each regression coefficient".

[0091] The bias term is used to calculate the linear predictor included in the logistic function. The bias term is a preset constant (for example, β).

[0092] Each regression coefficient includes a feeder regression coefficient as the first regression coefficient of the feeder, a head spindle regression coefficient as the first regression coefficient of the head spindle, a nozzle regression coefficient as the first regression coefficient of the nozzle, and a component size regression coefficient as the first regression coefficient of the component size. For example, if there are N f feeder units, N s head spindles, N n nozzles, and N c component sizes, there are corresponding numbers of the respective first regression coefficients. For example, if there are N f feeder units, the first regression coefficient is separately estimated for each of the N f feeder units. For example, the first regression coefficient of the feeder with the feeder serial number "FD00001" is α f [1], and the first regression coefficient of the feeder with the feeder serial number "FD00002" is α f [2]. The same applies to the head spindle, the nozzle, and the component size.

[0093] Referring again to Figure 1 , the linear regression coefficient estimation unit 24 is based on Figure 3The pre-processed data shown is used to estimate the second regression coefficient for estimating the tape feeding accuracy. The tape feeding accuracy indicates the positional accuracy when the feeder (tape feeder) supplies components to the component supply position by tape feeding.

[0094] The linear regression coefficient estimation unit 24 is based on the Figure 3 statistics (adsorption position offset statistics) obtained from the respective correction amounts shown, and the distribution of the measured values of the adsorption position offset (for example, the measured value of the tape feeding accuracy of the tape feeder), estimates the linear regression model that correlates the statistics and the measured value and its second regression coefficient (linear regression coefficient). For example, the linear regression coefficient estimation unit 24 plots the adsorption position offset statistics and the measured value of the feeding accuracy corresponding to the adsorption position offset statistics with the adsorption position offset statistics as the horizontal axis and the feeding accuracy (measured value of the inspection device) as the vertical axis, and estimates the second regression coefficient based on the plotted distribution. The inspection device is configured to include a hard sensor that directly measures the tape feeding accuracy. The linear regression coefficient estimation unit 24 is an example of the second coefficient estimation unit.

[0095] For example, the linear regression coefficient estimation unit 24 is based on the Figure 3 statistics in the X direction (X coordinate) (adsorption position offset statistics in the X direction) obtained from the correction amounts in the X direction among the respective correction amounts shown and the distribution of the measured values of the adsorption position offset in the X direction (for example, the measured value of the tape feeding accuracy in the X direction of the tape feeder), estimates the second regression coefficient in the X direction. Additionally, for example, the linear regression coefficient estimation unit 24 is based on the Figure 3 statistics in the Y direction (Y coordinate) (adsorption position offset statistics in the Y direction) obtained from the correction amounts in the Y direction among the respective correction amounts shown, and the distribution of the measured values of the adsorption position offset in the Y direction (for example, the measured value of the tape feeding accuracy in the Y direction of the tape feeder), estimates the second regression coefficient in the Y direction.

[0096] The second estimation model 34a generated by the linear regression coefficient estimation unit 24 is a simple regression model including the second regression coefficient, and this second regression coefficient represents the relationship between the tape feeding accuracy of each of the multiple units and the amount related to the adsorption position offset of the component.

[0097] Furthermore, the linear regression coefficient estimation unit 24 calculates the adsorption position offset statistics based on the recognition correction amount and the adsorption correction amount. For example, the linear regression coefficient estimation unit 24 calculates the adsorption position offset statistics based on the average value or median value of the recognition correction amount and the average value or median value of the adsorption correction amount. Additionally, the adsorption position offset statistics can also be calculated using the standard deviation. Also, the adsorption position offset statistics are calculated for each of the X coordinate and the Y coordinate. In other words, the second regression coefficient is estimated for each of the X coordinate and the Y coordinate.

[0098] The cause probability estimation device 30 outputs the cause of the mounting error and the estimation result of the tape feed accuracy based on the first regression coefficient and the second regression coefficient estimated by the coefficient estimation device 20, and the first mounting log L1. The first mounting log L1 is the log of the measurement object (estimation object) obtained during the production based on the mounting machine 10. The cause probability estimation device 30 outputs the cause of the mounting error and the estimation result of the tape feed accuracy using the first mounting log L1 obtained during the production based on the mounting machine 10 in parallel with the production based on the mounting machine 10. Thus, the cause probability estimation device 30 can assist in determining the abnormality of the unit by a manager or the like during the production of the mounting machine 10, and thus can perform maintenance based on a manager or the like before the unit fails.

[0099] The cause probability estimation device 30 includes: an error count preprocessing unit 31, a correction amount preprocessing unit 32, a mounting error cause estimation unit 33, a tape feed accuracy estimation unit 34, and summing units 35 to 38. The cause probability estimation device 30 can be implemented by a CPU, a memory, and the like. In addition, the processing of each functional block of the cause probability estimation device 30 is generally realized by a program execution unit such as a processor reading and executing software (program) recorded in a recording medium such as a ROM.

[0100] The functions of the error count preprocessing unit 31 and the correction amount preprocessing unit 32 are the same as those of the error count preprocessing unit 21 and the correction amount preprocessing unit 22, and the description thereof is omitted. The error count preprocessing unit 31 acquires the first mounting log L1 from the mounting machine 10, and generates the preprocessed data as shown in Figure 2 and outputs it to the mounting error cause estimation unit 33. In addition, the correction amount preprocessing unit 32 acquires the first mounting log L1 from the mounting machine 10, and generates preprocessed data including the data as shown in Figure 2 and the adsorption position offset statistic, and outputs it to the tape feed accuracy estimation unit 34.

[0101] The mounting error cause estimation unit 33 estimates the cause probability of the unit for the mounting error included in the preprocessed data based on the preprocessed data from the error count preprocessing unit 31 and the first regression coefficient from the non-linear regression coefficient estimation unit 23. The mounting error cause estimation unit 33 uses the first estimation model 33a based on the first regression coefficient to estimate the cause probability of the unit in the first mounting log L1 of the measurement object.

[0102] The mounting error cause estimation unit 33 calculates the probability that each factor (for example, each unit) becomes the cause for the generated error according to the binomial distribution derived from the preprocessed data and the first regression coefficient. As the binomial distribution model, the probability p of performing N adsorptions and generating y mounting errors is represented by the following formula 1.

[0103] [Mathematical formula 1]

[0104]

[0105] Among them, q represents the probability of installation error cause for each adsorption trial of each element. By transforming the above formula 1, the probability q of the installation error cause is represented by the following formula 2.

[0106]

[0107] Here, z is a linear predictor, and let α f be the first regression coefficient of the feeder, let α n be the first regression coefficient of the nozzle, let α s be the first regression coefficient of the head spindle, let α c be the first regression coefficient of the component size, let β be the bias term, and let chipW be the component size. Then this z is represented by the following formula 3.

[0108]

[0109] The linear predictor z can be calculated based on the pre-processed data from the error count pre-processing unit 31. Formulas 2 and 3 are an example of the first estimation model 33a. The first estimation model 33a is a statistical model based on the relationship between the installation error counts of multiple units and the abnormalities of each unit. The first estimation model 33a can also be said to be an identification model for identifying abnormal units from multiple units.

[0110] The installation error cause estimation unit 33 substitutes the linear predictor z obtained by substituting the first regression coefficients of the feeder, nozzle, head spindle, and component size when an installation error occurs into formula 2, thereby calculating the probability q of the installation error cause. The installation error cause estimation unit 33 calculates the probability q of the installation error cause for each of the feeder, nozzle, head spindle, and component size.

[0111] In addition, the probability q is the target variable in the first estimation model 33a. The first estimation model 33a is a multiple regression model with the first regression coefficients (partial regression coefficients) as explanatory variables and the probability q as the target variable.

[0112] The tape feed accuracy estimation unit 34 estimates the tape feed accuracy of the feeder of the mounting machine 10 based on the pre-processed data from the correction amount pre-processing unit 32 and the second regression coefficient from the linear regression coefficient estimation unit 24. A second estimation model 34a is generated including the second regression coefficient. The tape feed accuracy estimation unit 34 uses the second estimation model 34a based on the relationship between the tape feed accuracy of the tape feeder and the adsorption position offset statistic of the component to estimate the tape feed accuracy in the first mounting log L1 of the measurement object.

[0113] The tape feed accuracy estimation unit 34 uses the second regression coefficient to estimate the tape feed accuracy corresponding to the adsorption position offset statistic calculated based on the first mounting log L1 of the measurement object. The tape feed accuracy is, for example, a distance.

[0114] In addition, the tape feed accuracy estimation unit 34 further estimates the accuracy grade of the feeder based on a correspondence table between the tape feed accuracy and the accuracy grade (refer to the following Figure 5 ). This accuracy grade is the accuracy grade at this time point and can change over time.

[0115] Figure 5 FIG. is an example of a correspondence table showing the correspondence between the tape feed accuracy and the accuracy grade according to the present embodiment. The X-direction feed accuracy means the tape feed accuracy in the X direction, and the Y-direction feed accuracy means the tape feed accuracy in the Y direction.

[0116] As Figure 5 shown, in the correspondence table, for the X-direction feed accuracy and the Y-direction feed accuracy, one accuracy grade is established. In the Figure 5 example, the accuracy grades are "A", "B", "C", and "D", meaning that the feed accuracy decreases in this order.

[0117] The accuracy grade is an example of information based on the tape feed accuracy.

[0118] Referring again to Figure 1 , the totalization unit 35 accumulates the error cause probabilities for the component dimensions from the mounting error cause estimation unit 33 and outputs the error cause probabilities for each component serial number.

[0119] The totalization unit 36 accumulates the error cause probabilities for the nozzle from the mounting error cause estimation unit 33 and outputs the error cause probabilities for each nozzle serial number.

[0120] The totalization unit 37 accumulates the error cause probabilities for the head spindle from the mounting error cause estimation unit 33 and outputs the error cause probabilities for each head spindle number.

[0121] The totalizing unit 38 accumulates the error cause probability for the feeder from the installation error cause estimating unit 33 and the tape feeding accuracy from the tape feeding accuracy estimating unit 34, and outputs the error cause probability and the accuracy level for each feeder serial number.

[0122] The totalizing units 35 to 38 accumulate the error cause probabilities for a given production quantity (e.g., the quantity of one batch), and calculate one error cause probability for the production of the given production quantity based on the accumulated error cause probabilities. The totalizing units 35 to 38 may also, for example, sum up the error cause probabilities for the given production quantity and estimate the average value thereof as the one error cause probability. The error cause probability is an example of the estimation result.

[0123] The display device 40 displays various information. The display device 40 displays the error cause probability of the installation error and the estimation result of the tape feeding accuracy based on the cause probability estimating device 30. The display device 40 is a liquid crystal display device or the like, but is not limited thereto.

[0124] In addition, the coefficient estimating device 20, the cause probability estimating device 30, and the display device 40 may be implemented as separate devices, or at least two of them may be implemented as an integrated device. For example, the coefficient estimating device 20 and the cause probability estimating device 30 may be implemented as an integrated device, or the coefficient estimating device 20, the cause probability estimating device 30, and the display device 40 may be implemented as an integrated device.

[0125] [2. Operation of the Production Assistance System]

[0126] Next, with reference to Figures 6 to 11 the operation of the production assistance system 1 configured as described above will be described. Figure 6 is the first flowchart showing the operation (production assistance method) of the production assistance system 1 according to the present embodiment. Figure 6 shows the operation of estimating the first regression coefficient and the second regression coefficient. Figure 6 The processing of

[0127] As Figure 6 shown, the error count preprocessing unit 21 and the correction amount preprocessing unit 22 of the coefficient estimating device 20 respectively acquire the second installation log L2 for preliminary estimation (S10). The error count preprocessing unit 21 and the correction amount preprocessing unit 22 function as acquisition units for acquiring the second installation log L2.

[0128] Next, the error count preprocessing unit 21 and the correction amount preprocessing unit 22 respectively perform preprocessing (S20). The error count preprocessing unit 21, by performing preprocessing on the second installation log L2, generates, for example, Figure 2The pre-processed data shown is output to the non-linear regression coefficient estimation unit 23. The correction amount pre-processing unit 22 performs pre-processing on the second installation log L2, for example, to generate Figure 3 The pre-processed data shown is output to the linear regression coefficient estimation unit 24. In addition, in the present embodiment, the correction amount pre-processing unit 22 outputs the pre-processed data including the adsorption position offset statistic to the linear regression coefficient estimation unit 24.

[0129] Next, the non-linear regression coefficient estimation unit 23 estimates the first regression coefficient based on the pre-processed data from the error count pre-processing unit 21 (S30). Figure 7 is a diagram for explaining the estimation and use of the first regression coefficient according to the present embodiment. In addition, in Figure 7 the illustration of the error count pre-processing unit 21 and the error count pre-processing unit 31 is omitted.

[0130] As Figure 7 shown, the non-linear regression coefficient estimation unit 23 uses the MCMC method to estimate the first regression coefficient of each element based on the second installation log L2, and outputs the estimated first regression coefficient to the installation error cause estimation unit 33 of the cause probability estimation device 30. The installation error cause estimation unit 33 functions as an acquisition unit for acquiring the first regression coefficient.

[0131] Referring again to Figure 6 , next, the linear regression coefficient estimation unit 24 estimates the second regression coefficient based on the pre-processed data from the correction amount pre-processing unit 22 (S40). Figure 8 is a diagram for explaining the estimation and use of the second regression coefficient according to the present embodiment.

[0132] As Figure 8 shown, the linear regression coefficient estimation unit 24 estimates the second regression coefficient based on the adsorption position offset statistic based on the second installation log L2 and the distribution of the inspection measurement value M of the inspection device, and outputs the estimated second regression coefficient to the tape feed accuracy estimation unit 34 of the cause probability estimation device 30. The tape feed accuracy estimation unit 34 functions as an acquisition unit for acquiring the second regression coefficient.

[0133] In addition, Figure 8 the first installation log L1, the second installation log L2, and the inspection measurement value M shown are data obtained using the same installation machine 10.

[0134] Next, the process of performing the estimation of the production of the measurement object will be described. Figure 9 is a second flowchart showing the operation (production assistance method) of the production assistance system 1 according to the present embodiment. Figure 9The processing of steps S110 to S170 shown is executed by the cause probability estimation device 30, and step S180 is executed by the display device 40.

[0135] As Figure 9 shown, the error count preprocessing unit 31 and the correction amount preprocessing unit 32 of the cause probability estimation device 30 respectively acquire the first installation log L1 of the measurement object (S110). The error count preprocessing unit 31 and the correction amount preprocessing unit 32 function as acquisition units for acquiring the first installation log L1.

[0136] Next, the error count preprocessing unit 31 and the correction amount preprocessing unit 32 respectively perform preprocessing (S120). The error count preprocessing unit 31, by performing preprocessing on the first installation log L1, for example, generates Figure 2 the preprocessed data shown, and outputs it to the installation error cause estimation unit 33. In addition, the correction amount preprocessing unit 32, by performing preprocessing on the first installation log L1, for example, generates Figure 3 the preprocessed data shown, and outputs it to the tape feed accuracy estimation unit 34. In addition, in the present embodiment, the correction amount preprocessing unit 32 outputs the preprocessed data including the adsorption position offset statistic amount to the tape feed accuracy estimation unit 34.

[0137] Next, the installation error cause estimation unit 33 acquires the first regression coefficient from the non - linear regression coefficient estimation unit 23 (S130), and estimates the error cause probability of each element (S140). As Figure 7 shown, the installation error cause estimation unit 33 estimates the error cause probability of each element based on the preprocessed data from the error count preprocessing unit 31 (the preprocessed data based on the first installation log L1), the first regression coefficient, and the above - mentioned formulas 2 and 3. In Figure 7 it, an example of the installation error cause estimation unit 33 outputting the feeder cause probability is shown. The installation error cause estimation unit 33 outputs the error cause probability of each component serial number to the totalization unit 35, outputs the error cause probability of each nozzle serial number to the totalization unit 36, outputs the error cause probability of each head spindle number to the totalization unit 37, and outputs the error cause probability of each feeder serial number to the totalization unit 38.

[0138] Refer again to Figure 9 , next, the tape feed accuracy estimation unit 34 acquires the second regression coefficient from the linear regression coefficient estimation unit 24 (S150), and estimates the tape feed accuracy of the feeder (S160). As Figure 8As shown, the tape feeding accuracy estimation unit 34 estimates the tape feeding accuracy of the feeder based on the adsorption position offset statistic from the correction amount preprocessing unit 32 and the second regression coefficient. The tape feeding accuracy estimation unit 34 estimates the tape feeding accuracy in the X direction based on the adsorption position offset statistic in the X direction and the second regression coefficient in the X direction, and estimates the tape feeding accuracy in the Y direction based on the adsorption position offset statistic in the Y direction and the second regression coefficient in the Y direction.

[0139] Referring again to Figure 9 , next, the tape feeding accuracy estimation unit 34 estimates the accuracy level (S170) of the feeder based on the estimated tape feeding accuracies in the X direction and the Y direction and Figure 5 the corresponding table shown. The tape feeding accuracy estimation unit 34 outputs the accuracy level to the totalization unit 38.

[0140] Next, the display device 40 displays the information from each of the totalization units 35 to 38 (S180). Figure 10 FIG. is a diagram showing a first example of a screen displayed by the display device 40 according to the present embodiment. Figure 10 Shows the determination result of the installation error cause.

[0141] As Figure 10 shown, the display device 40 displays the error cause probability from the error cause probability estimation device 30. In Figure 10 , an example is shown in which the error cause probabilities for each of batches 01 to 03 for head spindles with addresses "1" and "2" are displayed. The address corresponds to the serial number. The probability of the cause of the installation error generated by the head spindle with address "1" in batch 01 is 10%, the probability of the cause of the installation error generated in batch 02 is 5%, and the probability of the cause of the installation error generated in batch 03 is 8%. In addition, the probability of the cause of the installation error generated by the head spindle with address "2" in batch 01 is 90%, the probability of the cause of the installation error generated in batch 02 is 80%, and the probability of the cause of the installation error generated in batch 03 is 100%. Thus, by displaying the error cause probabilities for each batch on one screen, the production support system 1 can assist the manager of the installation machine 10 in judging normal and abnormal factors. The cause probability is displayed, for example, for each unit, but is not limited thereto.

[0142] In addition, the display device 40 may also display the average value of the error cause probabilities for each batch ( Figure 10 the period average shown). Thereby, it is possible to more easily make a judgment on normal and abnormal factors by the manager.

[0143] In addition, for example, in the case of Lot 01, the total of the error cause probability of the head spindle at address "1" and the error cause probabilities of the nozzle, component, and feeder corresponding to the head spindle at address "1" is 100%.

[0144] Figure 11 FIG. 4 is a second example of a screen displayed by the display device 40 according to the present embodiment. Figure 11 Shows the determination result of the tape feed accuracy. Specifically, Figure 11 Shows the accuracy grade of the tape feed accuracy of the feeder ( Figure 11 the grade shown).

[0145] As Figure 11 shown, the display device 40 displays the determination result of the tape feed accuracy from the cause probability estimation device 30. In Figure 11 , an example is shown in which the determination results of the tape feed accuracies of each of Lots 01 to 03 at addresses "1" and "2" of the feeder are displayed. The address corresponds to the serial number. The accuracy grade of the feeder at address "1" (serial number: FD0001) in Lot 01 is "B", and the accuracy grades in Lots 02 and 03 are "A". In addition, the accuracy grade of the feeder at address "2" (serial number: FD0002) in Lot 01 is "D", the accuracy grade in Lot 02 is "C", and the accuracy grade in Lot 03 is "D". In this way, by displaying the accuracy grades of each lot of the feeder on one screen, the manager of the mounter 10 can be assisted in determining normal feeders and abnormal feeders.

[0146] In addition, when the display device 40 displays the estimation result of the feeder, the error cause probability and the accuracy grade may be on one screen. In addition, the display device 40 may display the estimated value of the tape feed accuracy instead of or together with the accuracy grade.

[0147] By estimating the error cause probability and the tape feed accuracy as described above, it is possible to estimate the state (the state of each unit) of the mounter 10 with high accuracy without increasing the cost (without adding a hard sensor). As a result, since accurate maintenance can be performed, equipment performance losses (for example, short-term or long-term stoppage of the mounter 10) can be reduced at low cost.

[0148] (Other Embodiments)

[0149] As described above, the production support system and the like according to one or more embodiments have been described based on the embodiments, but the present disclosure is not limited to these embodiments. As long as it does not deviate from the gist of the present disclosure, modes obtained by applying various modifications conceived by those skilled in the art to these embodiments and modes constructed by combining structural elements in different embodiments may also be included in the present disclosure.

[0150] For example, in the above-described embodiment, an example in which the installation error cause estimation unit estimates the cause probability of installation error by logistic regression analysis has been described. However, other non-linear regression analyses may also be used to estimate the cause probability of installation error. For example, the installation error cause estimation unit may also estimate the cause probability of installation error by polynomial regression analysis, support vector regression analysis, or the like.

[0151] In addition, in the above-described embodiment, an example in which the feeder is a belt feeder has been described. However, for example, it may also be other feeders such as a bulk feeder. In this case, the cause probability estimation device may not include a belt feeding accuracy estimation unit.

[0152] In addition, the coefficient estimation device and the cause probability estimation device according to the above-described embodiment may be terminal devices configured in a factory or the like equipped with an installation machine, or may be server devices configured away from the factory.

[0153] In addition, the first regression coefficient and the second regression coefficient according to the above-described embodiment may, for example, be estimated only once, or may be periodically estimated and the first regression coefficient and the second regression coefficient used in the cause probability estimation device may be periodically updated. In addition, the first regression coefficient and the second regression coefficient may, for example, also be estimated after a given event such as replacing a unit with a new one or repairing a failure has occurred.

[0154] In addition, the first installation log and the second installation log according to the above-described embodiment are also referred to as device logs.

[0155] In addition, in the above-described embodiment, an example in which the belt feeding accuracy is estimated based on the adsorption position offset statistic has been described. However, for example, the belt feeding accuracy may also be estimated based on either the recognition correction amount or the adsorption correction amount.

[0156] In addition, in the above-described embodiment and the like, each structural element may be constituted by dedicated hardware, or may be implemented by executing a software program suitable for each structural element. Each structural element may also be implemented by a program execution unit such as a CPU or a processor reading and executing a software program recorded in a recording medium such as a hard disk or a semiconductor memory.

[0157] In addition, the order of executing each step in the flowchart is illustrated for specifically explaining the present disclosure, and may be an order other than the above. In addition, a part of the above steps may be executed simultaneously (in parallel) with other steps, and a part of the above steps may not be executed.

[0158] In addition, the division of the functional blocks in the block diagram is an example. Multiple functional blocks can also be implemented as one functional block, one functional block can be divided into multiple ones, or a part of the function can be transferred to other functional blocks. Additionally, the functions of multiple functional blocks with similar functions can be processed in parallel or time-sharing by a single piece of hardware or software.

[0159] In addition, the production assistance system related to the above-described embodiments and the like can be implemented as a single device or can be implemented by multiple devices. In the case where the production assistance system is implemented by multiple devices, each structural element included in the production assistance system can also be arbitrarily assigned to the multiple devices. In the case where the production assistance system is implemented by multiple devices, the communication method between the multiple devices is not particularly limited and can be wireless communication or can be wired communication. Additionally, wireless communication and wired communication can also be combined between the devices.

[0160] In addition, each structural element described in the above-described embodiments and the like can be implemented as software or can typically be implemented as an integrated circuit, i.e., LSI. They can be individually made into a single chip or can include a part or all of them to be made into a single chip. Here, it is assumed to be LSI, but depending on the degree of integration, it is sometimes also called IC, system LSI, super LSI, or ultra LSI. In addition, the method of integrating into an integrated circuit is not limited to LSI and can also be implemented by a dedicated circuit (a general-purpose circuit that executes a dedicated program) or a general-purpose processor. An FPGA (Field Programmable Gate Array: field programmable gate array) that can be programmed after the LSI is manufactured or a reconfigurable processor that can reconfigure the connection or setting of the circuit units inside the reconfigurable LSI can also be used. Furthermore, if an integrated circuit technology that can replace LSI appears with the progress of semiconductor technology or other derived technologies, then of course, this technology can also be used for the integration of the structural elements.

[0161] A system LSI is a super-multi-functional LSI manufactured by integrating multiple processing units on one chip. Specifically, it is a computer system that includes a microprocessor, ROM (Read Only Memory: read-only memory), RAM (Random Access Memory: random access memory), etc. In the ROM, a computer program is stored. The microprocessor operates according to the computer program, and thus the system LSI achieves its function.

[0162] In addition, one aspect of the present disclosure can also be a computer program that causes a computer to execute Figure 6 and Figure 9 each of the characteristic steps included in the production assistance method shown in any one of them.

[0163] In addition, for example, the program can also be a program for causing a computer to execute. In addition, one aspect of the present disclosure can also be a non-transitory computer-readable recording medium storing such a program. For example, such a program can be recorded on a recording medium for distribution or circulation. For example, the distributed program can be installed on another device having a processor, and by causing the processor to execute the program, the device can perform the above-described various processes.

[0164] (Supplementary Note)

[0165] Through the description of the above embodiments, the following technologies are disclosed.

[0166] (Technology 1)

[0167] A production assistance system that estimates the probability that each of a plurality of units is a cause of an installation error occurring in an installation machine composed of the plurality of units. The production assistance system includes: an acquisition unit that acquires a first installation log of an object for estimating the probability from the installation machine, the first installation log including information related to the installation error; an installation error cause estimation unit that estimates the probability of each of the plurality of units based on the first installation log and a first estimation model, the first estimation model being a model based on the relationship between the number of installation errors of each of the plurality of units and the abnormality of the unit; and an output unit that outputs the estimation result of the installation error cause estimation unit. The first installation log includes information related to the production quantity and the number of installation errors among the plurality of units.

[0168] (Technology 2)

[0169] In the production assistance system according to Technology 1, the first estimation model is a multiple regression model. The multiple regression model uses the first regression coefficients of each of the plurality of units based on the distribution of the information related to the production quantity and the number of installation errors of each of the plurality of units as explanatory variables, and uses the probability as the target variable.

[0170] (Technology 3)

[0171] In the production assistance system according to Technology 2, the production assistance system further includes: a first coefficient estimation unit that estimates the first regression coefficients of each of the plurality of units based on the distribution of the information related to the production quantity and the number of installation errors included in a second installation log of the installation machine acquired before the first installation log.

[0172] (Technology 4)

[0173] In the production assistance system according to any one of Technologies 1 to 3, the output unit displays the probability of each of the plurality of units for the installation error.

[0174] (Technology 5)

[0175] In the production assistance system according to any one of Technologies 1 to 4, the plurality of units include a belt feeder that supplies components and a nozzle that adsorbs the components. The production assistance system further includes: a belt feed accuracy estimation unit that estimates the belt feed accuracy based on the first installation log and the second estimation model, where the second estimation model is a model based on the relationship between the belt feed accuracy of the belt feeder and the amount related to the adsorption position deviation of the components, and the first installation log further includes information related to the control amount for controlling the nozzle.

[0176] (Technology 6)

[0177] In the production assistance system according to Technology 5, the second estimation model is a simple regression model, and the simple regression model includes a second regression coefficient representing the relationship between the belt feed accuracy and the amount related to the adsorption position deviation of the components.

[0178] (Technology 7)

[0179] In the production assistance system according to Technology 6, the production assistance system further includes: a second coefficient estimation unit that estimates the second regression coefficient of each of the plurality of units based on the control amount for controlling the nozzle included in the third installation log of the installation machine obtained before the first installation log and the distribution of the measured values of the belt feed accuracy of the belt feeder.

[0180] (Technology 8)

[0181] In the production assistance system according to any one of Technologies 5 to 7, the output unit displays information related to the belt feed accuracy estimated by the belt feed accuracy estimation unit.

[0182] (Technology 9)

[0183] In the production assistance system according to any one of Technologies 1 to 8, the installation machine does not have a sensor that directly measures the states of the plurality of units.

[0184] (Technology 10)

[0185] In the production assistance system according to any one of Technologies 1 to 9, the plurality of units include a head spindle, a nozzle, and a feeder, and the installation error cause estimation unit estimates the probabilities of the head spindle, the nozzle, the feeder, and the components supplied by the feeder, respectively.

[0186] (Technology 11)

[0187] A production assistance method estimates the probability that each of a plurality of units is a cause of an installation error occurring in an installation machine composed of the plurality of units. In this production assistance method, a first installation log of an object for estimating the probability is obtained from the installation machine. The first installation log includes information related to the installation error. The probability of each of the plurality of units is estimated based on the first installation log and a first estimation model. The first estimation model is a model based on the relationship between the number of installation errors of each of the plurality of units and the abnormality of the unit. The estimated estimation result is output. The first installation log includes information related to the production quantity and the number of installation errors among the plurality of units.

[0188] (Technology 12)

[0189] A program for causing a computer to execute the production assistance method described in Technology 11.

[0190] Industrial applicability

[0191] This disclosure is useful in an assistance system or the like for assisting production using an installation machine.

[0192] Symbol description

[0193] 1 Production assistance system

[0194] 10 Installation machine

[0195] 20 Coefficient estimation device

[0196] 21, 31 Error count preprocessing unit

[0197] 22, 32 Correction amount preprocessing unit

[0198] 23 Nonlinear regression coefficient estimation unit (first coefficient estimation unit)

[0199] 24 Linear regression coefficient estimation unit (second coefficient estimation unit)

[0200] 30 Cause probability estimation device

[0201] 33 Installation error cause estimation unit

[0202] 33a First estimation model

[0203] 34 Tape feed accuracy estimation unit

[0204] 34a Second estimation model

[0205] 35, 36, 37, 38 Summation unit

[0206] 40 Display device

[0207] L1 First Installation Log

[0208] L2 Second Installation Log (Third Installation Log)

[0209] M Check the measured value.

Claims

1. A production assistance system that estimates the probability that each of a plurality of units is a cause of an installation error occurring in an installation machine composed of the plurality of units. The production assistance system includes: An acquisition unit that acquires a first installation log of an object for estimating the probability from the installation machine, the first installation log including information related to the installation error; An installation error cause estimation unit that estimates the probability for each of the plurality of units based on the first installation log and a first estimation model, the first estimation model being a model based on the relationship between the number of installation errors of each of the plurality of units and the abnormality of the unit; And An output unit that outputs the estimation result of the installation error cause estimation unit. The first installation log includes information related to the production quantity and the number of installation errors among the plurality of units.

2. The production assistance system according to claim 1, wherein The first estimation model is a multiple regression model, and the multiple regression model uses the first regression coefficient of each of the plurality of units based on the distribution of the information related to the production quantity and the number of installation errors of each of the plurality of units as an explanatory variable, and uses the probability as an objective variable.

3. The production assistance system according to claim 2, wherein The production assistance system further includes: a first coefficient estimation unit that estimates the first regression coefficient of each of the plurality of units based on the distribution of the information related to the production quantity and the number of installation errors included in a second installation log of the installation machine acquired before the first installation log.

4. The production assistance system according to any one of claims 1 to 3, wherein The output unit displays the probability for each of the plurality of units with respect to the installation error.

5. The production assistance system according to any one of claims 1 to 3, wherein The plurality of units include a belt feeder that supplies components and a nozzle that adsorbs the components. The production assistance system further includes: a belt feed accuracy estimation unit that estimates the belt feed accuracy based on the first installation log and a second estimation model, the second estimation model being a model based on the relationship between the belt feed accuracy of the belt feeder and the amount related to the deviation of the adsorption position of the component. The first installation log further includes information related to the control amount for controlling the nozzle.

6. The production assistance system according to claim 5, wherein The second estimation model is a simple regression model, and the simple regression model includes a second regression coefficient representing the relationship between the belt feed accuracy and the amount related to the deviation of the adsorption position of the component.

7. The production assistance system according to claim 6, wherein The production assistance system further includes: a second coefficient estimation unit that estimates the second regression coefficient of each of the plurality of units based on the distribution of the control amount for controlling the nozzle and the measured value of the belt feed accuracy of the belt feeder included in a third installation log of the installation machine acquired before the first installation log.

8. The production assistance system according to claim 5, wherein The output unit displays information related to the tape feed accuracy estimated by the tape feed accuracy estimation unit.

9. The production assistance system according to any one of claims 1 to 3, wherein, the mounting machine does not have a sensor for directly measuring the states of the plurality of units.

10. The production assistance system according to any one of claims 1 to 3, wherein, the plurality of units include a head spindle, a nozzle, and a feeder, the mounting error cause estimation unit estimates the probabilities of the head spindle, the nozzle, the feeder, and the component supplied by the feeder, respectively.

11. A production assistance method for estimating the probabilities that each of the plurality of units becomes a cause of a mounting error for a mounting error occurring in a mounting machine composed of a plurality of units, in the production assistance method, a first mounting log of an object for estimating the probability is obtained from the mounting machine, the first mounting log including information related to the mounting error, the probabilities of the plurality of units are estimated based on the first mounting log and a first estimation model, the first estimation model being a model based on the relationship between the number of mounting errors of each of the plurality of units and the abnormality of the unit, the estimated estimation result is output, the first mounting log includes information related to the production quantity and the number of mounting errors among the plurality of units.

12. A program for causing a computer to execute the production assistance method according to claim 11.

Images