WO2019013196A1

WO2019013196A1 - Manufacturing management device, manufacturing system, and manufacturing management method

Info

Publication number: WO2019013196A1
Application number: PCT/JP2018/026011
Authority: WO
Inventors: 太一清水; 博史天野; 多鹿　陽介; 裕一樋口
Original assignee: パナソニックＩｐマネジメント株式会社
Priority date: 2017-07-14
Filing date: 2018-07-10
Publication date: 2019-01-17

Abstract

A manufacturing management device (100) for managing a plurality of processes performed to manufacture a manufactured item (30). Each of the plurality of processes is performed using two or more constituent elements selected from a plurality of constituent elements. The manufacturing management device (100) is provided with: an acquisition unit (110) which acquires manufacturing log information indicating, with respect to each of the plurality of processes, two or more constituent elements used for a corresponding process and the results of the corresponding process; a calculation unit (120) which, by statistically processing the manufacturing log information, calculates the degree of contribution of each of the plurality of constituent elements to the manufacture of the manufactured item (30); and a display unit (130) which outputs the degree of contribution calculated by the calculation unit (120).

Description

Manufacturing management apparatus, manufacturing system and manufacturing management method

The present disclosure relates to a manufacturing management apparatus, a manufacturing system including the manufacturing management apparatus, and a manufacturing management method.

Conventionally, a component mounter includes a plurality of feeders and a plurality of nozzles, and a component supplied by a single feeder selected from a plurality of feeders is suctioned by a single nozzle selected from a plurality of nozzles. Mount on a board. At this time, there is known a technique in which the positional deviation accuracy of suction is detected for each nozzle or each feeder, and used to determine the necessity of maintenance (see, for example, Patent Document 1).

JP, 2013-98360, A

However, in the above-mentioned prior art, there is a problem that the component to be maintained is erroneously determined. For example, even if there is a large number of suction errors at a certain nozzle and it is determined that maintenance is necessary, in fact, a large number of suction errors may occur due to the influence of a feeder used simultaneously with the nozzle. As described above, there is a possibility that the position where it is determined that the maintenance of the feeder is required is erroneously determined as the maintenance of the nozzle.

When an erroneous determination occurs, the maintenance of the component that originally requires maintenance is not performed, and the abnormality remains unresolved. This leads to a decrease in production efficiency and quality of the product.

Thus, the present disclosure provides a manufacturing control apparatus, a manufacturing system, and a manufacturing control method that can support suppression of the reduction in production efficiency and quality of products.

In order to solve the above problems, a manufacturing management apparatus according to an aspect of the present disclosure is a manufacturing management apparatus that manages a plurality of processes performed to manufacture a product, and each of the plurality of processes is: The method is executed using two or more components selected from a plurality of components, and the manufacturing control device corresponds to each of the plurality of processes, the two or more components used in the corresponding process, and An acquisition unit for acquiring production log information indicating the results of the processing to be performed, and a calculation unit for calculating the contribution degree of each of the plurality of components to the production of the product by statistically processing the production log information And an output unit that outputs the degree of contribution calculated by the calculation unit.

Moreover, a manufacturing system according to an aspect of the present disclosure includes the manufacturing control apparatus, and at least one of the plurality of components, and includes a manufacturing facility that manufactures the product.

Further, a manufacturing control method according to an aspect of the present disclosure is a manufacturing control method for managing a plurality of processes performed to manufacture a product, wherein each of the plurality of processes includes a plurality of components. The method is executed using two or more selected components, and the manufacturing control method includes, for each of the plurality of processes, two or more components used in the corresponding process, and the results of the corresponding process By calculating the contribution of each of the plurality of constituent elements to the production of the product by calculating the production log information indicating the information and statistically processing the production log information, and the contribution calculated by the calculation unit Output

In addition, one aspect of the present disclosure can be realized as a program for causing a computer to function the manufacturing control method. Alternatively, it may be realized as a computer readable recording medium storing the program.

According to the present disclosure, it is possible to support reduction in production efficiency and quality of products.

FIG. 1 is a diagram showing the configuration of a manufacturing system according to the embodiment. FIG. 2 is a diagram illustrating an example of manufacturing log information acquired by the manufacturing management apparatus according to the embodiment. FIG. 3 is a block diagram showing the configuration of a manufacturing control apparatus according to the embodiment. FIG. 4 is a diagram showing input data of a logistic regression model by the manufacturing control apparatus according to the embodiment. FIG. 5 is a diagram showing a list of contribution degrees for each component calculated by the manufacturing control apparatus according to the embodiment. FIG. 6 is a flowchart showing the operation of the manufacturing control apparatus according to the embodiment. FIG. 7 is a diagram showing the number of occurrences of errors per predetermined number of times calculated by the manufacturing management apparatus according to the first modification of the embodiment. FIG. 8 is a diagram showing a list of contribution degrees for each component calculated by the manufacturing management apparatus according to the second modification of the embodiment. FIG. 9 is a diagram showing input data of a Poisson regression model by the manufacturing control apparatus according to the third modification of the embodiment. FIG. 10 is a diagram illustrating input data of a binomial logistic regression model by the manufacturing management apparatus according to the fourth modification of the embodiment.

(Summary of this disclosure)
In order to solve the above problems, a manufacturing control apparatus according to an aspect of the present disclosure is a manufacturing control apparatus that manages a plurality of processes performed to manufacture an article, and each of the plurality of processes is A plurality of components selected from a plurality of components, and the manufacturing control apparatus, for each of the plurality of processes, two or more components used in corresponding processes; Calculation for calculating the contribution of each of the plurality of components to the manufacture of the product by statistically acquiring the production log information indicating the corresponding processing performance and the production log information And an output unit that outputs the degree of contribution calculated by the calculation unit.

As a result, the degree of contribution is calculated for each component, and therefore the necessity of maintenance can be accurately determined based on the degree of contribution. For example, components that adversely affect manufacturing can be estimated based on the degree of contribution, and maintenance of the estimated bad components can be performed quickly. Therefore, according to the manufacturing control apparatus according to this aspect, it is possible to support the suppression of the deterioration of the production efficiency and the quality of the product.

In addition, for example, the calculation unit may calculate the degree of contribution by processing the manufacturing log information based on a generalized linear model.

Thus, by performing statistical processing based on the generalized linear model, it is possible to improve the reliability of the degree of contribution of each component. Specifically, the contribution of each component is the quantified effect of the corresponding component on manufacturing, and statistical processing based on the generalized linear model eliminates the effects of other components. ing. For this reason, components that adversely affect manufacturing can be accurately estimated based on the degree of contribution. As the estimation accuracy of the adversely affecting component is increased, it is not necessary for a person such as a maintenance worker or a manufacturing manager (operator) to investigate and judge the abnormal point, and to promptly cope with the bad component. Can.

Also, for example, the generalized linear model may be a logistic regression model, a Poisson regression model, or a binary logistic regression model.

Thereby, since the optimal regression model can be used according to the information classification of manufacture log information, the contribution degree with high reliability can be calculated.

Also, for example, the actual result is indicated by a flag indicating the presence or absence of an error in the corresponding processing, and the calculation unit calculates the contribution degree by processing the manufacturing log information based on a logistic regression model. It is also good.

As a result, since the occurrence of an error can be represented by a binary value of “0” and “1”, the contribution degree can be accurately calculated by using the logistic regression model.

Further, for example, the calculation unit may further calculate the number of predictions of an error that occurs when a predetermined component is used a predetermined number of times based on the degree of contribution of each of the plurality of components.

This makes it possible to present the presence or absence of the occurrence of an error in a more easily understandable manner to the maintenance worker or operator.

Also, for example, the output unit may be a display unit which graphically displays the degree of contribution of each of the plurality of components.

Thus, the degree of contribution can be presented in a display manner that can be easily understood by the maintenance worker or operator.

Also, for example, the plurality of components may be any of a plurality of manufacturing facilities for performing the plurality of processes, a plurality of component parts provided in each of the plurality of manufacturing facilities, and a plurality of components constituting the product. It may be

Thereby, not only the component part of a manufacturing installation but the contribution degree of the components which comprise a manufactured product can be calculated. Since the input data of statistical processing increases by increasing the calculation object of the degree of contribution, the degree of contribution can be calculated more accurately.

Also, for example, the plurality of components are classified into a plurality of component groups for each type, and each of the plurality of processes is executed using a component selected from each of the plurality of component groups. May be

As a result, since components can be managed for each type, bad components can be dealt with promptly.

In this way, the manufacturing control device helps to suppress the deterioration of the production efficiency and the quality of the product. For this reason, according to the manufacturing system according to the present aspect, it is possible to suppress the deterioration of the production efficiency and the quality of the product.

Further, a manufacturing control method according to an aspect of the present disclosure is a manufacturing control method for managing a plurality of processes performed to manufacture a product, wherein each of the plurality of processes includes a plurality of components. The method is executed using two or more selected components, and the manufacturing control method includes, for each of the plurality of processes, two or more components used in the corresponding process, and the results of the corresponding process By calculating the contribution of each of the plurality of constituent elements to the production of the product by calculating the production log information indicating the information and statistically processing the production log information, and the contribution calculated by the calculation unit May be output.

Thereby, similarly to the above-described manufacturing control apparatus, it is possible to support suppression of the reduction in production efficiency and quality of products.

Moreover, a program according to an aspect of the present disclosure is a program for causing a computer to execute the manufacturing method.

Embodiments will be specifically described below with reference to the drawings.

Note that all the embodiments described below show general or specific examples. Numerical values, shapes, materials, components, arrangement positions and connection forms of components, steps, order of steps, and the like shown in the following embodiments are merely examples, and are not intended to limit the present disclosure. Moreover, among the components in the following embodiments, components not described in the independent claims are described as optional components.

Further, each drawing is a schematic view, and is not necessarily illustrated exactly. Therefore, for example, the scale and the like do not necessarily match in each figure. Further, in each of the drawings, substantially the same configuration is given the same reference numeral, and overlapping description will be omitted or simplified.

Embodiment
[1. Constitution]
First, the configuration of a manufacturing management apparatus according to the first embodiment and a manufacturing system including the manufacturing management apparatus will be described with reference to FIG. FIG. 1 is a diagram showing the configuration of a manufacturing system 10 according to the present embodiment.

As shown in FIG. 1, the manufacturing system 10 includes a manufacturing facility 20 and a manufacturing control apparatus 100. In the manufacturing system 10 according to the present embodiment, the manufacturing facility 20 manufactures the product 30, and the manufacturing control apparatus 100 manages a plurality of processes performed to manufacture the product 30 by the manufacturing facility 20. .

The manufacturing facility 20 manufactures the product 30 by performing a plurality of processes. In the present embodiment, the manufacturing facility 20 is, for example, a component mounter. The product 30 has a substrate 31 and a plurality of components 32 mounted on the substrate 31.

In the present embodiment, the manufacturing facility 20 mounts the plurality of components 32 on the substrate 31. Specifically, the manufacturing facility 20 is an example of a manufacturing apparatus disposed on a manufacturing line of the product 30, and by mounting a plurality of parts 32 on each of a plurality of substrates 31 sequentially carried in, parts can be obtained. The substrate 31 with the 32 mounted thereon (ie, the product 30) is carried out. The carried-out substrate 31 (product 30) is transported to a manufacturing facility that performs the next manufacturing process (for example, a reflow process) or an inspection facility that performs an inspection of the product 30.

The production facility 20 includes a plurality of component groups each including a plurality of components (not shown) involved in the production of the product 30. The plurality of components include a feeder for supplying the component 32, a nozzle for suctioning the component 32, a header for holding the nozzle and moving between the feeder and the substrate 31 (lane in which the substrate 31 is transported). . For example, the manufacturing facility 20 includes a feeder group including a plurality of feeders, a nozzle group including a plurality of nozzles, a reel group including a plurality of reels, and a header group including a plurality of headers.

The product 30 is manufactured by performing a plurality of processes. The plurality of processes are, for example, individual mounting processes of the plurality of components 32. The plurality of processes may be performed simultaneously or sequentially.

Each of the plurality of processes is performed using two or more components selected from the plurality of components included in the manufacturing facility 20. In the present embodiment, the plurality of components also includes a component 32 which is an object to be mounted.

The manufacturing control apparatus 100 is an apparatus that manages a plurality of processes performed to manufacture the product 30. The manufacturing control apparatus 100 is, for example, a computer provided with a display or a computer connected to the display.

The manufacturing management apparatus 100 acquires manufacturing log information from the manufacturing facility 20, and manages a plurality of processes based on the acquired manufacturing log information. The manufacturing log information is data indicating the results of each of the plurality of processes performed by the manufacturing facility 20.

FIG. 2 is a diagram showing an example of manufacturing log information acquired by the manufacturing control apparatus 100 according to the present embodiment. As shown in FIG. 2, for each of a plurality of processes, the manufacturing log information indicates the time when the corresponding process was performed, the two or more components used in the corresponding process, and the results of the corresponding process. Is shown.

The time when the process is performed is, for example, at least one of the start time and the end time of the process. The start time and the end time are represented by, for example, a date indicated by year / month / day and a time indicated by hour: minute: second. The time may be expressed in units below second, such as milliseconds.

The processing results are indicated by a flag (error flag) indicating the presence or absence of an error in the corresponding processing. In the example shown in FIG. 2, when the error flag is “1”, it indicates that an error has occurred, and when the error flag is “0”, it indicates that an error has not occurred.

The plurality of processed components are managed, for example, for each component group. As shown in FIG. 2, the manufacturing facility 20 includes a unit A group, a unit B group, and a unit C group. The unit A group is a feeder group including a plurality of feeders (unit A). The unit B group is a nozzle group including a plurality of nozzles (unit B). The unit C group is a reel group composed of a plurality of reels (units C). Information indicated by an alphabet such as "A001", "B001" and "C001" and a three-digit number is an example of an identification number unique to each component. The way of assigning identification numbers is not particularly limited.

In the example shown in FIG. 2, the process P001 is performed using a feeder A having an identification number "A001", a nozzle B having an identification number "B001", and a reel C having an identification number "C001". Show that. In the following description, when “feeder A 001” is described, it means the feeder A whose identification number is “A 001”. The same applies to "nozzle B001" and "reel C001". Also, in FIG. 2, an identification number such as "P001" is given for each process, but this is described for the sake of clarity of the explanation, and it is not included in the manufacturing log information. Good.

In the present embodiment, one unit (feeder, nozzle and reel) is selected from each of unit A, unit B and unit C for each process, and the selected units cooperate with one another. Perform the corresponding processing. Note that depending on the type of processing, there may be units not selected. For example, certain processing may be performed by only two units, unit A and unit B. Also, multiple units may be selected from the same unit group. For example, another process may be performed by two or more units A.

Subsequently, the detailed configuration of the manufacturing control apparatus 100 will be described with reference to FIG. FIG. 3 is a block diagram showing the configuration of the manufacturing control apparatus 100 according to the present embodiment. As illustrated in FIG. 3, the manufacturing management apparatus 100 includes an acquisition unit 110, a calculation unit 120, a display unit 130, and a storage unit 140.

The acquisition unit 110 acquires manufacturing log information from the manufacturing facility 20. For example, the acquisition unit 110 acquires the manufacturing log information illustrated in FIG. 2 and stores the acquired manufacturing log information in the storage unit 140.

The acquisition unit 110 is, for example, a communication interface that communicates with the manufacturing facility 20. The communication may be either wireless communication or wired communication.

The calculation unit 120 statistically processes the manufacturing log information to calculate the contribution of each of the plurality of components to the manufacture of the product 30. The contribution of a component quantifies the influence of the component on manufacturing after excluding the influence of other components. Specifically, the contribution of the component corresponds to the degree of adverse effect on production, ie, the degree of badness.

Specifically, the calculation unit 120 calculates the degree of contribution by processing the manufacturing log information based on the generalized linear model. Generalized linear models include, but are not limited to, logistic regression models, Poisson regression models or binary logistic regression models.

In the present embodiment, the calculation unit 120 calculates the degree of contribution by processing the manufacturing log information based on the logistic regression model. Details of the process of calculating the degree of contribution based on the logistic regression model will be described later.

The calculation unit 120 may calculate the degree of contribution based on information in which at least one of the start time and the end time of the corresponding process is included in the predetermined aggregation period in the manufacturing log information. The aggregation period is a period during which the contribution of components used in the processing performed during the period is calculated, and is, for example, one hour to several hours, or one day to several days, etc. It is a period.

Note that by setting the aggregation period to a short period such as one minute to several minutes or one hour to several hours, it is possible to calculate the contribution degree with high real-time property. For this reason, abnormality of the component can be determined promptly based on the calculated degree of contribution, and manufacturing processes such as maintenance work such as member replacement can be improved. That is, it is not necessary to perform periodic maintenance and the like. Therefore, the downtime of the production line can be reduced and the production efficiency can be improved.

The display unit 130 is an example of an output unit that outputs the degree of contribution calculated by the calculation unit 120. In the present embodiment, the display unit 130 graphically displays the degree of contribution of each component. For example, the display unit 130 displays a list indicating the degree of contribution of each component.

The display unit 130 is, for example, a flat panel display such as a liquid crystal display (LCD) or an organic electroluminescence (EL) display, but is not limited thereto.

The storage unit 140 is a memory for storing the manufacturing log information acquired from the manufacturing facility 20, the calculated contribution degree, and the like. The storage unit 140 is a non-volatile memory such as a hard disk drive (HDD) or a semiconductor memory.

[2. Statistical processing]
Subsequently, statistical processing performed by the calculation unit 120 according to the present embodiment will be described.

In the present embodiment, the calculation unit 120 processes the manufacturing log information based on the logistic regression model. The logistic regression model is a regression model used when the dependent variable (target variable) is represented by two values. Specifically, the calculation unit 120 first generates input data of the logistic regression model based on the manufacturing log information.

FIG. 4 is a diagram showing input data of a logistic regression model by the manufacturing control apparatus 100 according to the present embodiment. In FIG. 4, each process is disposed on the vertical axis (column direction), and on the horizontal axis (row direction), a value y indicating the presence or absence of an error, and use and non-use of each component involved in the process. And a value x indicating. In the case shown by distinguishing values x for each component, as shown in FIG. 4, the value x is expressed as like x _{a1, x} _a2, x _b1 in accordance with the identification number of the components.

Also in FIG. 4, as in the case of the manufacturing log information shown in FIG. 2, an identification number such as "P001" is assigned to each process, but this is described to make the description easy to understand. , Not included in the input data.

The calculation unit 120 sets “0” or “1” to each of the value y indicating the presence or absence of an error and the value x indicating the use and non-use of all the components for each process based on the manufacturing log information. Assign a number.

Specifically, the calculation unit 120 assigns “1” to the value y corresponding to the process in which the error has occurred, and assigns “0” to the value y corresponding to the process in which the error has not occurred. That is, in FIG. 4, when the value y is “1”, it indicates that an error occurs in the corresponding process, and when the value y is “0”, no error occurs in the corresponding process. It is shown that.

In addition, the calculation unit 120 assigns “1” to the value x of the used component and assigns “0” to the value x of the unused component for each process. That is, in FIG. 4, when the component value x is “1”, it indicates that the component is used for the corresponding processing, and when the value x is “0”, the corresponding processing is performed. Indicates that the component has not been used.

For example, FIG. 4 shows that the feeder A 001 and the nozzle B 001 are used, the process P 001 is performed, and an error does not occur. Similarly, the feeder A 002 and the nozzle B 002 are used to perform the process P 002, indicating that an error has occurred.

The calculation unit 120 calculates the degree of contribution based on the logistic regression model, using the data shown in FIG. 4 as input data. In the logistic regression model, the error occurrence probability is represented by the following equation 1 with λ.

In Equation 1, C is a common constant term. x _a1 , x _a2 , x _b1 and x _b2 are values x of the respective components shown in FIG. 4. Each of a ₁ , a ₂ , b ₁ and b ₂ is a parameter (coefficient) of each component, and corresponds to the contribution of the component.

Logistic regression models follow the Bernoulli distribution. For this reason, the probability P (y = 1 | λ) that the actual value y is 1 is expressed by Equation 2 below.

Further, the probability P (y = 0 | λ) that the actual value y is 0 is expressed by the following Equation 3.

The calculation unit 120 substitutes the values of x and y into Equation 1 and Equation 2 for each process (row shown in FIG. 4), that is, for each combination of components, and P (y | λ) in the entire input data. Optimize parameters to maximize. As a result, the parameters of each component (specifically, a ₁ , a ₂ , b ₁ and b ₂ etc.), that is, the degree of contribution of each component are calculated.

FIG. 5 is a diagram showing a list of contribution degrees for each component calculated by the manufacturing control apparatus 100 according to the present embodiment. The degree of contribution indicates the inferiority of the component, so a higher numerical value indicates that an error is more likely to occur when the corresponding component is used.

In FIG. 5, the degree of contribution of each component is arranged in descending order. This indicates that the component positioned at the top (specifically, the feeder A 004) has the highest contribution, and the need for maintenance is high.

The list of contribution degrees shown in FIG. 5 is displayed on the display unit 130, for example. As a result, the necessity of maintenance can be easily determined for each component by looking at the list in which the maintenance worker or operator etc. are illustrated.

In the list, the degrees of contribution may be arranged in ascending order. Alternatively, they may be arranged in the ascending or descending order of the identification numbers of the components.

[3. Operation]
Subsequently, an operation (manufacturing management method) of the manufacturing management apparatus 100 according to the present embodiment will be described with reference to FIG. FIG. 6 is a flowchart showing the operation of the manufacturing control apparatus 100 according to the present embodiment.

As shown in FIG. 6, first, the acquisition unit 110 acquires manufacturing log information from the manufacturing facility 20 (S10). For example, the acquiring unit 110 acquires, for each predetermined period such as one hour to several hours or one day to several days, manufacturing log information on the process performed in the period. Alternatively, the acquiring unit 110 may acquire manufacturing log information on the process each time the manufacturing facility 20 performs the process. The acquisition unit 110 stores the acquired manufacturing log information in the storage unit 140.

Next, the calculation unit 120 calculates the degree of contribution for each component (S20). Specifically, as described above, the calculation unit 120 calculates the contribution degree for each component based on the logistic regression model using the input data shown in FIG. 4. The calculation unit 120 calculates, for example, the degree of contribution for each predetermined period, such as one hour or one day.

The display unit 130 graphically displays the calculated degree of contribution (S30). For example, as shown in FIG. 5, the display unit 130 displays the degree of contribution of each component in a list.

As described above, according to the manufacturing management apparatus 100 according to the present embodiment, the degree of contribution is calculated for each component by performing analysis based on the regression model. The contribution of a component is based on the contribution because the influence of the corresponding component on manufacturing is quantified and statistical processing based on the regression model excludes the influence of other components. Thus, the necessity of maintenance can be determined accurately. For example, components that adversely affect manufacturing can be estimated based on the degree of contribution, and maintenance of the estimated bad components can be performed quickly.

For example, a person such as a maintenance worker or a production manager (operator) does not have to investigate and judge the abnormal part, and bad components can be dealt with promptly. Therefore, according to the manufacturing control apparatus 100 which concerns on this Embodiment, suppression of the fall of the productive efficiency of the product 30, and quality can be assisted.

(Modification 1)
Subsequently, a first modification of the embodiment will be described.

The manufacturing management apparatus according to the present modification calculates the number of predicted errors that may occur for each component based on the calculated degree of contribution, and displays the calculation result. In the present modification, the configurations of the manufacturing control apparatus 100 and the manufacturing system 10 are the same as those of the embodiment, and thus the description thereof is omitted.

In the present modification, the calculation unit 120 calculates the number of times of error that occurs when the corresponding component is used a predetermined number of times. The predetermined number of times is, for example, 10000 times, but is not limited to this, and may be 1000 times or 100,000 times.

Specifically, since the parameter for each component, that is, the degree of contribution is calculated based on the equation 1 and the equation 2, the calculation unit 120 uses the degree of contribution as a known value Calculate the number of errors predicted for each element. Specifically, the calculation unit 120 calculates the number of predicted errors (error occurrence probability) p when the target component is used once based on Equation 4 below.

In Equation 4, the contribution of the average unit is a value obtained by averaging the contributions of one or more units that can be used in combination with the target unit. For example, when the target unit is the feeder A 001, the contribution of the average unit is a value obtained by averaging the contributions of all the nozzles that can be combined with the feeder A 001.

The calculation unit 120 calculates the number of occurrences of errors per 10000 times by multiplying the calculated p by 10000. The calculation unit 120 generates, for example, a list of the number of times of errors illustrated in FIG. 7 by calculating 10000 × p for all the constituent elements. FIG. 7 is a diagram showing the number of times of occurrence of an error in 10000 trials for each component calculated by the manufacturing management apparatus 100 according to the present modification.

As a result, the presence or absence of an error can be displayed in a more easily understandable display manner for a maintenance worker or an operator.

FIG. 7 shows the result of calculating the degree of contribution on a daily basis. Thereby, the temporal change of the degree of contribution can also be represented.

(Modification 2)
Next, a second modification of the embodiment will be described.

In the embodiment described above, as shown in FIG. 5, the numerical values of the degree of contribution are illustrated as a list. On the other hand, in the present modification, the degree of contribution is shown graphically by means other than numerical values. In the present modification, the configurations of the manufacturing control apparatus 100 and the manufacturing system 10 are the same as those of the embodiment, and thus the description thereof is omitted.

Specifically, the display unit 130 may classify the degree of contribution into a plurality of ranges according to the value as shown in the list shown in FIG. Here, FIG. 8 is a view showing another example of the list of contribution degrees for each component calculated by the manufacturing management apparatus 100 according to the present embodiment.

The degree of contribution is classified into, for example, four ranges according to the value. In FIG. 8, predetermined colors are determined in advance for each range of contribution degree. For example, the smaller the degree of contribution, the whiter, and the larger the degree of contribution, the bluer. In FIG. 8, the difference in color is expressed by the difference in density of dots. Note that the density difference of dots as shown in FIG. 8 or the type of hatching may be different for each range, instead of the color.

For example, the calculation unit 120 performs division and setting of the range of possible values of the degree of contribution by performing clustering with the calculated degree of contribution for each component as input data. The clustering method is, for example, a Ward method or a k-means method, but is not limited thereto. Further, the number of divisions of the range may not be four, and may be two or more. In addition, the calculation unit 120 may equally divide the range of the degree of contribution.

Further, in FIG. 8, the calculation unit 120 calculates the contribution degree over a plurality of periods, and the results are displayed together. Specifically, the calculation unit 120 classifies the manufacturing log information according to the day on which the process is performed based on the start date and time of the process, and calculates the degree of contribution on a daily basis. In FIG. 8, Day 1 to Day 4 indicate the days on which the process was performed.

As described above, according to the present modification, the degree of contribution can be presented in a manner that can be easily understood by the maintenance worker or the operator.

As in the first modification, the number of errors per predetermined number may be displayed by color coding.

(Modification 3)
Then, the modification 3 is demonstrated.

In this modification, compared with the embodiment, the manufacturing log information acquired by the manufacturing management apparatus 100 is different, and along with this, the regression model used by the calculating unit 120 is different. The configurations of the manufacturing control apparatus 100 and the manufacturing system 10 are the same as those of the embodiment, and thus the description thereof is omitted.

In the present modification, the calculation unit 120 processes the manufacturing log information based on the Poisson regression model. The Poisson regression model is a regression model used when the dependent variable (target variable) is a number value (count data) without an upper limit number.

FIG. 9 is a diagram showing input data of a Poisson regression model by the manufacturing control apparatus 100 according to the present modification. In addition, in FIG. 9, the combination of the used component is shown similarly to the manufacture log information shown in FIG.

As shown in FIG. 9, in the Poisson regression model, the producible number per day is used as the actual value y, instead of the error flag. The number of producible products is a count value (count data) substantially without an upper limit.

In the present modification, the calculation unit 120 calculates the degree of contribution based on the Poisson regression model, using the data shown in FIG. 9 as input data. In the Poisson regression model, the average number of predicted production is represented by λ, and is represented by the following Equation 5.

In Equation 5, C is a common constant term. x _a1 , x _a2 , x _b1 and x _b2 are values x of the respective components shown in FIG. 4. Each of a ₁ , a ₂ , b ₁ and b ₂ is a parameter (coefficient) of each component, and corresponds to the contribution of the component. These are the same as in the embodiment. The left side of Equation 5 may be only λ instead of log λ.

The Poisson regression model follows the Poisson distribution. Therefore, the probability P (y | λ) is expressed by the following Equation 6.

The calculation unit 120 substitutes the values of x and y into Equations 5 and 6 for each combination of components (rows shown in FIG. 9), and maximizes P (y | λ) in the entire input data. To optimize the parameters. As a result, the parameters of each component (specifically, a ₁ , a ₂ , b ₁ and b ₂ etc.), that is, the degree of contribution of each component are calculated.

In the present modification, the larger the actual value y, the larger the number of producible products, which is preferable as the manufacturing equipment 20. Therefore, the contribution of the component corresponds to the degree of good influence on production, ie, the degree of goodness.

As described above, according to the manufacturing management apparatus 100 according to the present modification, the contribution degree for each component can be calculated even when the number value with no upper limit is used as the actual value. The contribution of a component is based on the contribution because the influence of the corresponding component on manufacturing is quantified and statistical processing based on the regression model excludes the influence of other components. Thus, the necessity of maintenance can be determined accurately. Thereby, similarly to the embodiment, it is possible to support the suppression of the reduction of the production efficiency and the quality of the product 30.

(Modification 4)
Subsequently, Modification 4 will be described.

In this modification, as in the third modification, compared with the embodiment, the manufacturing log information acquired by the manufacturing management apparatus 100 is different, and accordingly, the regression model used by the calculation unit 120 is different. The configurations of the manufacturing control apparatus 100 and the manufacturing system 10 are the same as those of the embodiment, and thus the description thereof is omitted.

In the present modification, the calculation unit 120 processes the manufacturing log information based on the binomial logistic regression model. The binomial logistic regression model is a regression model used when the dependent variable (target variable) is an upper limit number having a number value (count data).

FIG. 10 is a diagram showing input data of a binomial logistic regression model by the manufacturing management apparatus 100 according to the present modification. In addition, in FIG. 10, the combination of the used component is shown similarly to the manufacture log information shown in FIG.

As shown in FIG. 10, in the binomial logistic regression model, the number of times of occurrence of an error y and the number of times of implementation N are used as the actual value y, instead of the error flag. Note that the number of mountings corresponds to the number of processes performed in the corresponding combination. The number of occurrences of errors y does not exceed the number of implementation times N. That is, y is a count value (count data) whose upper limit is N.

In the present modification, the calculation unit 120 calculates the degree of contribution based on a binomial logistic regression model, using the data shown in FIG. 10 as input data. In the binomial logistic regression model, the occurrence probability of an error is represented by the following Equation 7 as q.

In Equation 7, C is a common constant term. x _a1 , x _a2 , x _b1 and x _b2 are values x of the respective components shown in FIG. 4. Each of a ₁ , a ₂ , b ₁ and b ₂ is a parameter (coefficient) of each component, and corresponds to the contribution of the component. These are the same as in the embodiment.

A binomial logistic regression model follows a binomial distribution. Therefore, the probability P (y | N, q) is expressed by the following equation 8.

The calculation unit 120 substitutes the values of x, y and N into Equations 7 and 8 for each combination of components (rows shown in FIG. 10), and P (y | N, q) is the entire input data. Optimize parameters for maximum. As a result, the parameters of each component (specifically, a ₁ , a ₂ , b ₁ and b ₂ etc.), that is, the degree of contribution of each component are calculated.

As described above, according to the manufacturing management apparatus 100 according to the present modification, the contribution degree for each component can be calculated even when the upper limit value is used as the actual value. The contribution of a component is based on the contribution because the influence of the corresponding component on manufacturing is quantified and statistical processing based on the regression model excludes the influence of other components. Thus, the necessity of maintenance can be determined accurately. Thereby, similarly to the embodiment, it is possible to support the suppression of the reduction of the production efficiency and the quality of the product 30.

In the present variation, for example, the degree of contribution for each component lot or manufacturer of parts is calculated with respect to the number of failures of electronic devices such as smart phones. This makes it possible to estimate lots with high failure rates or manufacturers.

(Other embodiments)
The manufacturing control apparatus, the manufacturing system, and the manufacturing control method according to one or more aspects have been described above based on the embodiments, but the present disclosure is not limited to these embodiments. As long as various modifications that can occur to those skilled in the art are made to the present embodiment, and forms configured by combining components in different embodiments are included within the scope of the present disclosure, without departing from the gist of the present disclosure. Be

For example, the regression model and probability distribution shown in the above embodiment and modification are merely examples, and other regression models and probability distributions may be used. For example, normal distribution or gamma distribution may be used.

Also, for example, in the above embodiment, a list of contribution degrees is displayed, but among the calculated contribution degrees, only contribution degrees and components (specifically, constituent elements that adversely affect) that are larger than the threshold are You may display it.

For example, although the display unit is illustrated as an example of the output unit that outputs the degree of contribution, the output unit may be an audio output unit. For example, the output unit may output a component with a high degree of contribution as audio data. Alternatively, the output unit may print a list of contribution degrees on a medium such as paper.

Also, for example, although an example in which the manufacturing facility 20 is a component mounter has been shown, the present invention is not limited to this. For example, the manufacturing equipment 20 may be a processing device for processing a raw material such as metal or resin, and a molding device for molding a processed material.

Further, the component may not be any one of the manufacturing equipment, the components provided in the manufacturing equipment, and the parts constituting the product. Specifically, the components do not have to be physically present. For example, the component may be a condition relating to manufacturing, which is an element that may affect production performance. Specifically, the component may be the condition of equipment such as the speed of a head or a part suction method, or may be the condition of a person such as a worker in charge or a group in charge. Alternatively, the component may be conditions of the production method such as the presence or absence of splicing, or environmental conditions such as season.

Further, (1) Each of the above-described devices may be specifically a computer system including a microprocessor, a ROM, a RAM, a hard disk unit, a display unit, and the like. A computer program is stored in a RAM (Ramdom Access Memory) or a hard disk unit. Each device achieves its function by the microprocessor operating according to the computer program. Here, the computer program is configured by combining a plurality of instruction codes indicating instructions to the computer in order to achieve a predetermined function.

(2) A part or all of the components constituting each of the above-described devices may be configured from one system LSI (Large Scale Integration: large scale integrated circuit). The system LSI is a super-multifunctional LSI manufactured by integrating a plurality of components on one chip, and specifically includes a microprocessor, a ROM (Read Only Memory), a RAM, etc. Computer system. A computer program is stored in the RAM. The system LSI achieves its functions by the microprocessor operating according to the computer program.

(3) A part or all of the components constituting each of the above-described devices may be composed of an IC card or a single module which can be detached from each device. The IC card or module is a computer system including a microprocessor, a ROM, a RAM, and the like. The IC card or module may include the above-described ultra-multifunctional LSI. The IC card or module achieves its functions by the microprocessor operating according to the computer program. This IC card or this module may be tamper resistant.

(4) The present disclosure may be the method described above. In addition, it may be a computer program that realizes these methods by a computer, or may be a digital signal composed of a computer program.

(5) The present disclosure relates to a computer program or a recording medium capable of reading digital signals from a computer, such as a flexible disk, a hard disk, a CD-ROM, an MO, a DVD, a DVD-ROM, a DVD-RAM, a BD (Blu-ray (registration It may be recorded on a trademark (trademark) Disc), a semiconductor memory or the like. In addition, digital signals recorded on these recording media may be used.

(6) The present disclosure may transmit a computer program or a digital signal via a telecommunication line, a wireless or wired communication line, a network represented by the Internet, data broadcasting, and the like.

(7) The present disclosure is a computer system provided with a microprocessor and a memory, the memory storing the computer program, and the microprocessor may operate according to the computer program.

(8) It may be implemented by another independent computer system by recording and transferring the program or digital signal on a recording medium, or by transferring the program or digital signal via a network or the like .

Further, various modifications, replacements, additions, omissions and the like can be made in the above-described embodiments within the scope of the claims or the equivalent thereof.

The present disclosure can be used as, for example, a production management device that can help suppress the production efficiency and the deterioration of product quality, and can be used, for example, for management of production in a factory.

DESCRIPTION OF SYMBOLS 10 Manufacturing system 20 Manufacturing equipment 30 Manufacture 31 Substrate 32 Parts 100 Manufacturing management apparatus 110 Acquisition part 120 Calculation part 130 Display part 140 Storage part

Claims

A manufacturing control apparatus for managing a plurality of processes performed to manufacture a product, comprising:
Each of the plurality of processes is performed using two or more components selected from a plurality of components,
The manufacturing control device is
An acquisition unit configured to acquire, for each of the plurality of processes, manufacturing log information indicating two or more components used in the corresponding process and the results of the corresponding process;
A calculation unit that calculates contribution of each of the plurality of components to the production of the product by statistically processing the production log information;
And an output unit that outputs the degree of contribution calculated by the calculation unit.
The manufacturing management apparatus according to claim 1, wherein the calculation unit calculates the contribution degree by processing the manufacturing log information based on a generalized linear model.
The manufacturing control apparatus according to claim 2, wherein the generalized linear model is a logistic regression model, a Poisson regression model, or a binary logistic regression model.
The result is indicated by a flag indicating the presence or absence of an error in the corresponding processing,
The manufacturing management apparatus according to claim 3, wherein the calculation unit calculates the contribution degree by processing the manufacturing log information based on a logistic regression model.
The calculation unit further calculates, based on the degree of contribution of each of the plurality of components, the number of times of prediction of an error that occurs when a predetermined component is used a predetermined number of times. The manufacturing control device according to item 1.
The manufacturing control apparatus according to any one of claims 1 to 5, wherein the output unit is a display unit that graphically illustrates and displays the degree of contribution of each of the plurality of components.
The plurality of components are any of a plurality of manufacturing facilities for performing the plurality of processes, a plurality of component parts provided in each of the plurality of manufacturing facilities, and a plurality of components constituting the product. The manufacturing control device according to any one of items 1 to 6.
The plurality of components are classified into a plurality of component groups by type.
The manufacturing control apparatus according to any one of claims 1 to 7, wherein each of the plurality of processes is executed using a component selected from each of the plurality of component groups.
A manufacturing control apparatus according to any one of claims 1 to 8;
A manufacturing system comprising: at least one of the plurality of components; and a manufacturing facility for manufacturing the product.
A manufacturing control method for managing a plurality of processes performed to manufacture a product, comprising:
Each of the plurality of processes is performed using two or more components selected from a plurality of components,
The manufacturing control method is
For each of the plurality of processes, manufacturing log information indicating two or more components used in the corresponding process and the results of the corresponding process is acquired;
Calculating the contribution of each of the plurality of components to the production of the product by statistically processing the production log information;
A manufacturing control method that outputs the calculated contribution.
A program for causing a computer to execute the manufacturing control method according to claim 10.