JP2018195130A - Abnormal cause estimating apparatus, abnormal cause estimating method, and abnormal cause estimating program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an improved abnormality cause estimating device, which quantifies the magnitude relationship and the degree of deviation between process control data estimated as an abnormality cause and process control data in a normal state, and presents it as information on the direction and degree of abnormality. A cause estimation method and an abnormality cause estimation program are provided. An abnormality cause estimation apparatus according to an aspect of the present disclosure obtains all process tracking data from an input device, and standard value data including at least one of a standard upper limit value, a standard lower limit value, and a center value thereof. , Process tracking reference data is set from all process tracking data based on the standard deviation of the process control data value of a predetermined inspection item, and a process is performed based on at least one of a predetermined product type and production lot from all process tracking data. The tracking analysis data is set, the variation of the analysis data with respect to the reference data is calculated for each process control item, and a graph showing the variation is drawn on the display device. [Selection diagram] Figure 1

Description

  The present disclosure relates to an abnormality cause estimation device, an abnormality cause estimation method, and an abnormality cause estimation program.

  As one of conventional methods for estimating the cause of abnormality in a production process such as a device manufacturing process or a material process process, for example, a method of diagnosing the cause of abnormality by pattern recognition using Mahalanobis distance is known (patent) Reference 1).

  In the configuration described in Patent Document 1, a Mahalanobis space is calculated from a steady state of a diagnosis target, a feature amount (process management data) is extracted from the target to be diagnosed, a Mahalanobis distance is calculated, and a preset threshold value is calculated. And the Mahalanobis distance are compared. Next, when the Mahalanobis distance is larger than the threshold value, the maximum value of the distance element value is specified, and the feature amount corresponding to the maximum value of the distance element value is replaced with the average value of the feature amounts of the steady state data. After that, a new Mahalanobis distance is calculated using the replaced feature amount, and the above is repeated until the newly calculated Mahalanobis distance is less than or equal to the threshold value, and the effect of the feature amount of all process control items is drawn in a bar graph. The In the bar graph, the process management item corresponding to the long bar is diagnosed as the cause of the abnormality.

JP 2004-227279 A

  However, in the configuration of the conventional Patent Document 1, the magnitude relationship (direction of abnormality) and the degree of deviation (degree of abnormality) between the process management data estimated as the cause of abnormality and the process management data in the normal state are unknown. . For this reason, there is a problem that information on the direction and degree of abnormality is insufficient, and it is difficult to make an appropriate abnormality countermeasure plan based on the diagnosis.

  The purpose of this disclosure is an improved anomaly cause estimation that quantifies the magnitude relationship and degree of divergence between the process control data estimated as the cause of anomaly and the normal state process management data, and presents it as information on the direction and extent of the anomaly An apparatus, an abnormality cause estimation method, and an abnormality cause estimation program are provided.

  An abnormality cause estimation device according to an aspect of the present disclosure includes all process tracking data including process management data values for a target product type and production lot from an input device, and a standard that defines a range of the process management data values that satisfy the standard A data acquisition unit for acquiring standard value data including at least one of an upper limit value and a standard lower limit value and a standard central value that is a central value thereof, and the process management data value of a predetermined inspection item from all the process tracking data Based on the standard deviation, the process management data value is selectively acquired and set as reference data, and the process set in the reference data for each process management item based on the reference data and the standard value data A reference data setting unit for setting process tracking reference data consisting of values obtained by normalizing the management data values, and a predetermined product type and production from all the process tracking data. Based on at least one of the lots, the process management data value is selectively acquired and set as analysis data, and is included in the analysis data for each process management item based on the analysis data and the standard value data. An analysis data setting unit that sets process tracking analysis data composed of values obtained by normalizing the process management data value, a variation calculation unit that calculates a variation of the analysis data with respect to the reference data for each of the process management items, and And a graph drawing unit that draws a graph indicating fluctuation on a display device.

  An abnormality cause estimating method according to an aspect of the present disclosure includes a process tracking total data including a process management data value for a target product type and a production lot from an input device, and a standard that defines a range of the process management data value that satisfies the standard. A process tracking all data acquisition step for acquiring standard value data including at least one of an upper limit value and a standard lower limit value and a standard center value that is a center value thereof, and the process of a predetermined inspection item from the process tracking all data Based on the standard deviation of the management data value, the process management data value is selectively obtained and set as reference data. Based on the reference data and the standard value data, the process management items are set in the reference data. In addition, a process tracking reference data setting process for setting process tracking reference data consisting of values obtained by normalizing the process management data values, and the process tracking reference data , Based on at least one of a predetermined product type and production lot, the process management data value is selectively acquired and set as analysis data, and based on the analysis data and the standard value data, for each process management item, A process tracking analysis data setting step for setting process tracking analysis data composed of values obtained by normalizing the process management data values included in the analysis data, and a change in the analysis data with respect to the reference data for each of the process management items. A configuration is provided that includes a variation calculation step for calculating, and a graph drawing step for drawing a graph indicating the variation on a display device.

  An abnormality cause estimation program according to an aspect of the present disclosure includes, on a computer, all process tracking data including process management data values for a target product type and a production lot from an input device, and a range of the process management data values satisfying a standard. A process tracking all data acquisition step for acquiring standard value data including at least one of a standard upper limit value and a standard lower limit value and a standard central value that is a central value thereof, and predetermined inspection items from the process tracking total data Based on the standard deviation of the process management data value, the process management data value is selectively obtained and set as reference data, and the reference data is set for each process management item based on the reference data and the standard value data. A process tracking reference data setting step for setting process tracking reference data consisting of values obtained by normalizing the process management data values set to From all the process tracking data, based on at least one of a predetermined product type and production lot, the process management data value is selectively obtained and set as analysis data, and based on the analysis data and the standard value data, For each process management item, a process tracking analysis data setting step for setting process tracking analysis data consisting of values obtained by normalizing the process management data values included in the analysis data; and for each process management item, the reference data A variation calculation step for calculating the variation of the analysis data and a graph drawing step for drawing a graph indicating the variation on a display device are employed.

  According to the present disclosure, improved abnormality cause estimation that quantifies the magnitude relationship and the degree of deviation between the process management data estimated as the cause of abnormality and the normal state process management data and presents it as information on the direction and degree of abnormality. An apparatus, an abnormality cause estimation method, and an abnormality cause estimation program can be provided.

It is a block diagram of the abnormality cause estimation system which concerns on 1st Embodiment. It is a flowchart which shows the operation | movement flow of the abnormality cause estimation apparatus which concerns on 1st Embodiment. It is a flowchart which shows the operation | movement flow of step S1150 of the flowchart shown in FIG. It is an example of all the process tracking data. It is an example of standard value data. It is a flowchart which shows the operation | movement flow of step S2200 of the flowchart shown in FIG. It is a flowchart which shows the operation | movement flow of step S3200 of the flowchart shown in FIG. It is a flowchart which shows the operation | movement flow of step S4200 of the flowchart shown in FIG. It is an example of the normalized reference data and analysis data. It is a flowchart which shows the operation | movement flow of step S5250 of the flowchart shown in FIG. It is an example of variation factor diagram data. It is a flowchart which shows the operation | movement flow of step S6200 of the flowchart shown in FIG. It is an example of the process tracking chart with respect to several analysis data. It is an example of the process tracking chart with respect to single analysis data. It is an example of a fluctuation factor diagram. It is a figure which shows an example of the hardware constitutions of a computer.

  Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings.

(First embodiment)
FIG. 1 is a configuration diagram of an abnormality cause estimation system 100 according to the first embodiment. The abnormality cause estimation system 100 includes an input device 110, an abnormality cause estimation device 120, and a display device 130.

  The input device 110 generates all process tracking data. Here, the entire process tracking data is data obtained by collecting characteristic values (process management data values) indicating the state in the process corresponding to each process management item for the target product type and lot. In one example, varieties and lots include those in progress. The feature value is, for example, a measurement value acquired from a sensor such as a temperature sensor, a pressure sensor, or an image sensor provided in the process, a value indicating a change amount of the measurement value, and an average of the measurement value over a predetermined time. The average value. In addition, the measured value etc. which were provided in the process or input with the input terminal which an operator carries are also considered. The input device 110 is, for example, a temperature sensor, a pressure sensor, an image sensor, a computer connected to these sensors, or an input terminal.

  The input device 110 further generates standard value data. Here, the standard value data is data defining a range of process management data values satisfying the standard for each process management item. In one example, the standard value data includes a standard upper limit value, a standard central point, and a standard central value that define an upper limit value, a central value, and a lower limit value of a range of values that satisfy the standard.

  The abnormality cause estimation device 120 includes a data acquisition unit 121, a standard data creation unit 122, a reference data setting unit 123, an analysis data setting unit 124, a fluctuation calculation unit 125, and a graph drawing unit 126. In one example, the abnormality cause estimating device 120 includes a CPU and a ROM, and the CPU includes a computer that reads and executes a program stored in the ROM.

  The data acquisition unit 121 acquires all process tracking data from the input device 110 and arranges data values of all process tracking data. Details of the processing contents will be described later with reference to FIGS. 3 and 4A. Furthermore, the data acquisition unit 121 acquires standard value data. Details of the processing contents will be described later with reference to FIGS. 3 and 4B.

  The standard data creation unit 122 creates process tracking standard data based on all process tracking data and standard value data, and arranges data values of the process tracking standard data. Here, the process tracking standard data is included in all the process tracking data so that, for each process management item, the standard upper limit value is 1, the standard center value is 0.5, and the standard lower limit value is 0, for example. This is data consisting of values normalized by linearly converting feature values. Here, normalization means performing a transformation such as dividing the representative value by a representative value in order to compare the target value with other values. Details of the processing contents will be described later with reference to FIGS. 5A and 6.

  The reference data setting unit 123 selects, based on the entire process tracking data and the standard value data, data selected and acquired based on the standard deviation of the process management data value of a predetermined inspection item from all the process tracking data. Set as. In one example, the reference data setting unit 123 obtains a function value corresponding to the standard deviation value of the process management data value of a predetermined inspection item, using a predefined function having a standard deviation value in the domain. Data selected and acquired based on the obtained function value is set as reference data.

  The reference data setting unit 123 further sets process tracking reference data based on the reference data and the standard value data, and arranges data values of the process tracking reference data. Here, the process tracking reference data is data composed of values obtained by normalizing process management data values set in the reference data for each process management item. In one example, the process management data value is 1 so that the process management data value equal to the standard upper limit value is 1, the process management data value equal to the standard center value is 0.5, and the process management data value equal to the standard lower limit value is 0. Linearized and normalized. In another example, the process management data value is such that the process management data value equal to the standard upper limit value is 1, the process management data value equal to the standard center value is 0, and the process management data value equal to the standard lower limit value is -1. Are linearly transformed and normalized. Details of the processing contents will be described later with reference to FIGS. 5B and 6.

  The analysis data setting unit 124 sets analysis data based on all process tracking data and standard value data. Here, the analysis data is data composed of one or more records selected from all the process tracking data using, for example, the product type and the production lot as key values. The analysis data setting unit 124 further sets process tracking analysis data based on the analysis data and the standard value data, and arranges data values of the process tracking analysis data. Here, the process tracking analysis data is data composed of values obtained by normalizing the process management data values set in the analysis data for each process management item. In one example, the process management data value is 1 so that the process management data value equal to the standard upper limit value is 1, the process management data value equal to the standard center value is 0.5, and the process management data value equal to the standard lower limit value is 0. Linearized and normalized. In another example, the process management data value is such that the process management data value equal to the standard upper limit value is 1, the process management data value equal to the standard center value is 0, and the process management data value equal to the standard lower limit value is -1. Are linearly transformed and normalized. Details of the processing contents will be described later with reference to FIGS. 5C and 6.

  The fluctuation calculation unit 125 calculates the fluctuation of the analysis data with respect to the reference data for each process management item. Details of the processing contents will be described later with reference to FIGS.

  The graph drawing unit 126 draws a graph indicating fluctuation. Details of the processing contents will be described later with reference to FIG. 9 and FIGS.

  The display device 130 displays the analysis graph drawn by the graph drawing unit 126. The display device 130 is, for example, a liquid crystal display, an organic EL display, a printer, or a plotter.

  FIG. 2 is a flowchart showing an operation flow of the abnormality cause estimation device 120 according to the first embodiment. This process is realized, for example, by the CPU of the abnormality cause estimating device 120 reading and executing a program stored in the ROM.

  In step S1150, the abnormality cause estimation device 120 acquires all process tracking data (processing as the data acquisition unit 121). Details of the processing content will be described later with reference to FIGS. 3, 4A, and 4B.

  In step S2200, the abnormality cause estimating device 120 creates process tracking standard data (processing as the standard data creating unit 122). Details of the processing contents will be described later with reference to FIGS. 5A and 6.

  In step S3200, the abnormality cause estimation device 120 sets process tracking reference data (processing as the reference data setting unit 123). Details of the processing contents will be described later with reference to FIGS. 5B and 6.

  In step S4200, the abnormality cause estimating apparatus 120 sets process tracking analysis data (processing as the analysis data setting unit 124). Details of the processing contents will be described later with reference to FIGS. 5C and 6.

  In step S5250, the abnormality cause estimation device 120 calculates the variation of the analysis data with respect to the reference data (processing as the variation calculation unit 125). Details of the processing contents will be described later with reference to FIGS.

  In step S6200, the abnormality cause estimation device 120 draws an analysis graph (processing as the graph drawing unit 126). Details of the processing contents will be described later with reference to FIGS. 9, 10A, 10B, and 10C.

  FIG. 3 is a flowchart showing an operation flow of step S1150 of the flowchart shown in FIG. FIG. 4A is an example of all process tracking data. FIG. 4B is an example of standard value data.

  In step S1000, the data acquisition unit 121 acquires process management data for all processes including all types including work in progress and all production lots.

  In step S1050, the data acquisition unit 121 sets, for all process management data, values corresponding to process management numbers and process management item names as columns, values corresponding to product types and production lots as rows, and in ascending order of process management numbers. The process management data values are arranged in a matrix format in which the columns are arranged from left to right, and the rows are arranged from top to bottom in ascending order of product types and production lots.

  As shown in FIG. 4A, the first row in the matrix format is the process management number 7000, and the second row is the process management item name 7050. Also, the first column of the matrix format is the product type 7100 and the second column is the production lot 7150. All the process tracking data is created by arranging the process management data value 7200 by arranging the rows in the ascending order of the production processes and from the top to the bottom in the ascending order of the types and production lots.

  In step S1100, the data acquisition unit 121 acquires the standard upper limit value, the standard center value, and the standard lower limit value of the process management items for all processes.

  Then, as shown in FIG. 4B, the data acquisition unit 121, for example, has a drawing category 8000, a drawing number 8050, a process management number 8100, a process management item name 8150, a standard type 8200, a standard upper limit value, a standard center value, a standard Standard value data in which a lower limit value 8250 and a provisional value width 8300 are arranged is created.

  Here, the drawing section 8000 is, for example, a variable that takes two values of 0 and 1, and the initial value is 1. The drawing number 8050 is a serial number having 1 as an initial value. The standard type 8200 is a variable that takes three values of 0, 1, and 2. A value of 1 indicates that the standard value data has no standard lower limit value, and a value of 2 indicates that the standard value data does not have a standard upper limit value. A value of 3 indicates that the standard value data includes both a standard upper limit value and a standard lower limit value.

  The provisional value width 8300 is a coefficient α for calculating a provisional standard value when there is no standard value. For example, α = 0.1. When there is no standard upper limit value (the value of the standard type 8200 is 2), the standard center value × (1 + α) is set as the provisional standard upper limit value. When there is no standard lower limit value (the value of the standard type 8200 is 1), the standard center value × (1−α) is set as the provisional standard lower limit value. When there is no standard center value, for example, (standard upper limit value + standard lower limit value) / 2 is set as the provisional standard center value.

  5A, FIG. 5B, and FIG. 5C are flowcharts showing operation flows of step S2200, step S3200, and step S4200 of the flowchart shown in FIG. 2, respectively. FIG. 6 is an example of normalized reference data and analysis data.

  First, FIG. 5A and FIG. 6 will be referred to. In step S2200, the standard upper limit value is normalized to 1, the standard center value is normalized to 0.5, and the standard lower limit value is normalized to 0 for each process management item.

  In step S2050, for each process management item, the drawing number, the presence / absence of the standard, the process management number, the process management item name, the normalized standard upper limit value, and the normalized standard lower limit value are set in ascending order of the process management number. Arrange from left column to right column. The tracking Min 9150 shown in FIG. 6 is a normalized standard lower limit value and is equal to zero. The tracking Max 9200 is a normalized upper limit value that is equal to 1.

  Reference is now made to FIGS. 5B and 6.

  In step S3000, data having the process capability index Cpk of the final process inspection item of 1.33 or more is selected and acquired from the process management data of all processes, and the reference data is set.

  In step S3050, the reference data is normalized with the standard value for each process management item. In step S3100, a maximum value, a minimum value, and an average value are calculated for the normalized reference data for each process management item. In step S3150, for each process management item, the drawing number, the presence / absence of a standard, the process management number, the process management item name, the maximum value, the minimum value, and the average value of normalized reference data are set in ascending order of the process management number. Arrange from left column to right column.

  The normalized reference data Min, reference data Max, and reference data Ave 9250 shown in FIG. 6 represent the maximum value, the minimum value, and the average value of the normalized reference data, respectively.

  Reference is now made to FIGS. 5C and 6. In step S4000, process management data corresponding to a predetermined product type and production lot number is selected and acquired from the process management data of all processes, and set to one or more analysis data.

  In step S4050, the analysis data is normalized with a standard value for each process management item. In step S4100, a maximum value, a minimum value, and an average value are calculated for the normalized analysis data for each process management item. In step S4150, the drawing number, the presence / absence of the standard, the process management number, the process management item name, the maximum value, the minimum value, and the average value of the normalized analysis data are changed from the left column to the right column in ascending order of the process management number. Arrange.

  The normalized analysis data Min, analysis data Max, and analysis data Ave9300 shown in FIG. 6 respectively represent the maximum value, the minimum value, and the average value of the normalized analysis data.

  When step S4200 shown in FIG. 2 is executed, as shown in FIG. 6, a drawing number 9000, standard presence / absence 9050, process management number, process management item name 9100, normalized tracking Min 9150, tracking Max 9200 Normalized data for a process tracking chart in which normalized reference data Min, Max, Ave 9250 and normalized analysis data Min, Max, Ave 9300 are arranged is created.

  FIG. 7 is a flowchart showing an operation flow of step S5250 of the flowchart shown in FIG.

  In step S5000, the fluctuation calculation unit 125 compares the average value x of the normalized analysis data with the average value M0 of the normalized reference data for each process management number. Depending on the comparison result, the process proceeds to step S5050 (when x> M0), step S5100 (when x = M0), or step S5150 (when x <M0). Next, in steps S5050, S5100, and S5150, the variation Δ is set according to the following formula (1).

ここで、Ｍ_ｍａｘ、Ｍ_ｍｉｎは、それぞれ、基準データの最大値および最小値を表す。

Here, M _max and M _min represent the maximum value and the minimum value of the reference data, respectively.

  Next, in step S5200, the fluctuation calculation unit 125 arranges the presence / absence of the standard, the process management number, the process management item name, and the fluctuation from the left column to the right column in ascending order of the process management numbers.

  FIG. 8 is an example of variation factor diagram data. As shown in FIG. 8, normalization data for a variation factor diagram in which standard presence / absence 10000, process management number, process management item name 10050, and variation 10100 are arranged is created.

  FIG. 9 is a flowchart showing an operation flow of step S6200 of the flowchart shown in FIG. FIG. 10A is an example of a process tracking chart for a plurality of analysis data. FIG. 10B is an example of a process tracking chart for a single analysis data. FIG. 10C is an example of a variation factor diagram. In these process tracking charts and variation factor diagrams, data of process management items whose drawing section flag value is 1 are drawn from the left column to the right column in ascending order of the drawing numbers.

  In step S6000, the graph drawing unit 126 assigns a drawing number and a drawing classification flag (0 or 1) to each process management item. A value of 1 indicates that drawing is performed, and a value of 0 indicates that drawing is not performed. In one example, the value of the drawing classification flag is set by the user via a user interface (not shown) provided in the abnormality cause estimation device 120.

  In step S6050, the graph drawing unit 126 determines the value of the drawing division flag. If the value of the drawing division flag is 0, drawing is not necessary, and step S6200 is terminated. On the other hand, if the value of the drawing classification flag is 1, the process advances to step S6110.

  In step S6110, the graph drawing unit 126 draws the normalized standard upper limit value, standard center value, and standard lower limit value for each process management item. In one example, as shown in FIGS. 10A and 10B, the process management item name 10500 and the presence / absence of a standard 11500 (* mark) are displayed as items in the axial direction in which the process management items are arranged. The normalized standard upper limit value and standardized lower limit value are drawn as a straight line 11000 having a value in the axial direction indicating a standard value of 1 and a straight line 11450 having a value of 0, respectively. Further, in one example, the normalized standard center value is drawn as a straight line (not shown) having a direction value indicating the standard value of 0.5.

  In step S6120, the graph drawing unit 126 draws the maximum value, the minimum value, and the average value of the normalized reference data for each process management item. In one example, as shown in FIGS. 10A and 10B, the maximum value 11100, the minimum value 11400, and the average value 11250 of the normalized reference data are each drawn as a line graph. A region surrounded by the maximum value 11100 and the minimum value 11400 is drawn as a band graph 11150.

  In step S6130, the graph drawing unit 126 draws the maximum value, the minimum value, and the average value of the normalized analysis data for each process management item. In an example, when there are a plurality of pieces of analysis data, as shown in FIG. 10A, the maximum value 11200, the minimum value 11350, and the average value 11300 of the normalized analysis data are each drawn as a line graph. When the analysis data is single, as shown in FIG. 10B, the average value 11300 of the normalized analysis data is drawn as a line graph.

  In step S6150, the graph drawing unit 126 draws the variation Δ of the analysis data with respect to the reference data for each process management item. As shown in FIG. 10C, the fluctuation is drawn as a bar graph for the process management item whose drawing section flag value is 1. Here, the process management item name 10500 and the standard presence / absence 11500 (indicated by *) are displayed as items in the axial direction in which the process management items are arranged.

  In one example, the graph drawing unit 126 draws a graph indicating the maximum value, the minimum value, and the average value of the reference data and a graph indicating the maximum value, the minimum value, and the average value of the analysis data with different colors.

  In one example, one or more thresholds for ranking the degree of variation divergence are set. For example, when variation Δ ≦ 0.5, the divergence degree is ranked into the best rank ○. When 0.5 <variation Δ ≦ 1.0, the divergence degree is ranked into an intermediate rank Δ. When 1.0 <variation Δ, the degree of divergence is ranked in the worst rank ▲. The bars 13000 indicating the variations belonging to the best rank ◯, the intermediate rank Δ, and the worst rank ▲ are shown in different colors, for example, green, yellow, and red.

  Further, the graph drawing unit 126 (see FIG. 1) may draw the process tracking charts of FIGS. 10A and 10B and the variation factor diagram of FIG. 10C so as to be arranged vertically. Thereby, the user can have a bird's-eye view of these charts and the whole figure via the display device 130 (see FIG. 1).

  As described above, the abnormality cause estimation device 120 according to the present disclosure defines the range of process tracking data including the process management data value for the target product type and production lot from the input device 110 and the range of the process management data value satisfying the standard. A data acquisition unit 121 for acquiring standard value data including at least one of a standard upper limit value and a standard lower limit value and a standard central value that is a central value thereof, and process management data of a predetermined inspection item from all process tracking data Based on the standard deviation of the value, the process management data value is selectively acquired and set as reference data, and the process management data value set in the reference data is set for each process management item based on the reference data and the standard value data. Based on the reference data setting unit 123 for setting the process tracking reference data based on all the process tracking data, at least a predetermined product type and production lot Process management data values are selected and acquired as analysis data based on the process data, and the process tracking is performed for each process management item based on the process management data value included in the analysis data based on the analysis data and standard value data. An analysis data setting unit 124 that sets analysis data, a variation calculation unit 125 that calculates a variation of analysis data with respect to reference data for each process management item, a graph drawing unit 126 that draws a graph indicating the variation on a display device, The structure provided with is taken.

  In an industrial production process, there are processes such as a device manufacturing process and a material process process in which many processes are connected, and an upstream manufacturing process affects the performance of a downstream manufacturing process. The abnormality cause estimation method and the abnormality cause estimation program according to the present disclosure provide the material conditions, manufacturing conditions, workmanship, facility conditions, and manufacturing conditions obtained from the production process when a defect occurs in such a production process. Analyze process management data such as environmental conditions. By analyzing the process management data, the directionality and degree of abnormality of the process management data which is the cause of the defect can be estimated well. As described above, the abnormality cause estimation method and the abnormality cause estimation program according to the present disclosure can be applied to the use of abnormality cause estimation.

  Furthermore, according to the abnormality cause estimation method and the abnormality cause estimation program according to the present disclosure, the overall magnitude and the degree of divergence between the abnormal state process management data and the normal state process management data are qualitatively and quantitatively viewed. be able to. Further, the magnitude relationship and the degree of divergence between the process management data estimated as the cause of the abnormality and the process management data in the normal state can be quantified, and information on the directionality and degree of abnormality can be provided. Therefore, the user can plan an appropriate abnormality countermeasure based on the provided information.

  FIG. 11 is a diagram illustrating an example of a hardware configuration of a computer. The function of each part in each embodiment and each modification described above is realized by a program executed by the computer 2100.

  As shown in FIG. 11, a computer 2100 includes an input device 2101 such as an input button and a touch pad, an output device 2102 such as a display and a speaker, a CPU (Central Processing Unit) 2103, a ROM (Read Only Memory) 2104, and a RAM (Random Access Memory) 2105 is provided. The computer 2100 also reads information from a recording medium such as a hard disk device, a storage device 2106 such as an SSD (Solid State Drive), a DVD-ROM (Digital Versatile Disk Read Only Memory), or a USB (Universal Serial Bus) memory. 2107, a transmission / reception device 2108 that performs communication via a network is provided. Each unit described above is connected by a bus 2109.

  Then, the reading device 2107 reads the program from a recording medium on which a program for realizing the functions of the above-described units is recorded, and stores the program in the storage device 2106. Alternatively, the transmission / reception device 2108 communicates with the server device connected to the network, and causes the storage device 2106 to store a program for realizing the function of each unit downloaded from the server device.

  Then, the CPU 2103 copies the program stored in the storage device 2106 to the RAM 2105, and sequentially reads out and executes the instructions included in the program from the RAM 2105, thereby realizing the functions of the above-described units. Further, when executing the program, the RAM 2105 or the storage device 2106 stores information obtained by various processes described in each embodiment, and is used as appropriate.

(Other embodiments)
In the first embodiment, sequential numbers of 1, 2, 3,. Instead, the process management number includes a management item number and a management item classification number in the process. If the process management classification number is a causal system such as manufacturing conditions, work contents, and environment, 0, inspection result, etc. An embodiment in which the result system is 1 is also conceivable.

  In the first embodiment, 10% is set as the provisional value width 8300. Instead, an embodiment in which the provisional value width 8300 is variable is also conceivable.

  In the first embodiment, a provisional value width calculation formula based on the count α is provided as the provisional standard value. Instead of this, an embodiment in which the provisional standard upper limit value, standard center value, and standard lower limit value are set based on reference data set in the past is also conceivable. For example, the average value + 3σ (σ represents a standard deviation) is set as a provisional standard upper limit value, the average value −3σ is set as a provisional standard lower limit value, and the average value is set as a provisional standard center value. Further, by setting a coefficient other than 3 as the coefficient of σ, it is possible to cope with various customer requested quality levels.

  In the first embodiment, the standard upper limit value, the standard center value, and the standard lower limit value normalized by setting the standard upper limit value to 1, the standard center value to 0.5, and the standard lower limit value to 0 are provided. Instead of this, an embodiment in which a standard upper limit value, a standard center value, and a standard lower limit value normalized by setting the standard upper limit value as 1, the standard center value as 0, and the standard lower limit value as -1 can be considered.

  In the first embodiment, an inspection item for the final process is provided as a predetermined inspection item for selecting reference data. Instead of this, an embodiment in which a predetermined inspection item for selecting reference data is used as an inspection item for an intermediate process is also conceivable.

  In the first embodiment, a condition that the process capability index Cpk is 1.33 or more is provided as a condition of the process management data value that is selectively acquired and set in the reference data. Instead of this, an embodiment in which the process capability index Cpk is changed to a condition of 1.67 or more is also conceivable. Furthermore, instead of this, an embodiment in which the deviation value is not included in the standard deviation range (that is, not included in the average value ± σ range) is also conceivable. As described above, by changing the conditions for selecting and acquiring the reference data, it is possible to meet various quality requirements of customers.

  In the present embodiment, the straight lines corresponding to the normalized standard upper limit value, standard center value, and standard lower limit value are represented as straight lines having axis values indicating standard values of 1, 0.5, and 0, respectively. Draw with. Instead of this, the straight line corresponding to the normalized standard upper limit value, standard center value, and standard lower limit value is drawn as straight lines with axis values indicating standard values of 1, 0, and −1, respectively. Forms are also conceivable.

  The abnormality cause estimation apparatus, the abnormality cause estimation method, and the abnormality cause estimation program according to the present disclosure are connected in many processes, such as a device manufacturing process and a material process process, and an upstream manufacturing process is a downstream manufacturing process. It can be applied to production processes that affect the workmanship of

DESCRIPTION OF SYMBOLS 100 Abnormal cause estimation system 110 Input device 120 Abnormal cause estimation apparatus 121 Data acquisition part 122 Standard data creation part 123 Reference data setting part 124 Analysis data setting part 125 Fluctuation calculation part 126 Graph drawing part 130 Display apparatus

Claims (20)

All process tracking data including process management data values for the target product type and production lot from the input device, and standard upper and lower limit values that define the range of the process management data values that satisfy the standard, and the standard value thereof A data acquisition unit for acquiring standard value data including at least one of the central value;
Based on the standard deviation of the process management data value of a predetermined inspection item from all the process tracking data, the process management data value is selectively acquired and set as reference data, and based on the reference data and the standard value data A reference data setting unit for setting process tracking reference data composed of values obtained by normalizing the process management data values set in the reference data for each process management item;
From all the process tracking data, based on at least one of a predetermined product type and production lot, the process management data value is selectively acquired and set as analysis data, and based on the analysis data and the standard value data, For each process management item, an analysis data setting unit that sets process tracking analysis data composed of values obtained by normalizing the process management data values included in the analysis data;
For each process management item, a fluctuation calculation unit that calculates fluctuations of the analysis data with respect to the reference data;
A graph drawing unit for drawing a graph indicating the fluctuation on a display device;
An abnormality cause estimation device comprising: The data acquisition unit
The process management number corresponding to the process management item and the value corresponding to the process management item are set as a column, the value corresponding to the product type and the production lot is set as a row, and the column is changed from left to right in ascending order of the process control number. The abnormality cause estimation device according to claim 1, wherein the process management data values are arranged in a matrix format in which the rows are arranged from top to bottom in ascending order of the product type and production lot.   If the standard upper limit value is not included in the acquired standard value data, the data acquisition unit sets the provisional standard upper limit value based on the value of the standard center value, and the acquired standard value data includes the standard value upper limit value. When the standard lower limit value is not included, the provisional standard lower limit value is set based on the standard center value, and when the acquired standard value data does not include the standard central value, the standard upper limit value and The abnormality cause estimation device according to claim 1, wherein the provisional standard center value is set based on the standard lower limit value.   The data acquisition unit sets a provisional standard upper limit value, a provisional standard lower limit value, or a provisional standard center value based on an average value and standard deviation of the reference data set in the past. Item 10. The abnormality cause estimation device according to Item 1.   The reference data setting unit and the analysis data setting unit are configured such that the process management data value equal to the standard upper limit value is 1, the process management data value equal to the standard center value is 0.5, and the standard lower limit value is equal to the standard lower limit value. The abnormality cause estimation device according to claim 1, wherein the process management data value is normalized so that the process management data value is zero.   The reference data setting unit and the analysis data setting unit are configured such that the process management data value equal to the standard upper limit value is 1, the process management data value equal to the standard center value is 0, and the process management equal to the standard lower limit value. The abnormality cause estimation device according to claim 1, wherein the process management data value is normalized so that the data value is −1. The reference data setting unit includes:
For each process control item, calculate the maximum value, minimum value, and average value of the process tracking reference data,
2. The process management number, the process management item name, the maximum value, the minimum value, and the average value are arranged from left column to right column in ascending order of process management numbers corresponding to the process management items. The abnormality cause estimation device described in 1. The analysis data setting unit
For each process management item, calculate the maximum value, minimum value, and average value of the process tracking analysis data,
2. The process management number, the process management item name, the maximum value, the minimum value, and the average value are arranged from left column to right column in ascending order of process management numbers corresponding to the process management items. The abnormality cause estimation device described in 1.   The abnormality cause estimation device according to claim 1, wherein the predetermined inspection item is an inspection item in an intermediate process.   The abnormality cause estimation device according to claim 1, wherein the predetermined inspection item is an inspection item of a final process.   2. The abnormality cause estimation device according to claim 1, wherein the process management data value selected and acquired and set in the reference data is the process management data value having a process capability index value of 1.33 or more.   The abnormality cause estimation device according to claim 1, wherein the process management data value selected and acquired and set in the reference data is the process management data value having a process capability index value of 1.67 or more.   The abnormality cause estimation device according to claim 1, wherein the process management data value selected and acquired and set in the reference data is the process management data value whose deviation value is not included in the range of the standard deviation. The variation calculator is
For each process control item, the average value x of the normalized analysis data value is compared with the average value M0 of the normalized reference data,
The variation Δ is expressed by the following formula (1):

The abnormality cause estimation device according to claim 1, wherein M _max and M _min represent a maximum value and a minimum value of the reference data, respectively. The graph drawing unit
In ascending order of drawing numbers corresponding to the process management items, the process management items corresponding to the drawing numbers are displayed from the left column to the right column,
Draw a straight line corresponding to the standardized upper limit value, the standard center value, and the standard lower limit value,
Draw a graph showing the maximum, minimum, and average values of the normalized reference data,
Draw a graph showing the maximum value, minimum value, and average value of the normalized analysis data,
The abnormality cause estimation device according to claim 1, wherein a graph showing the fluctuation is drawn.   The abnormality cause estimation device according to claim 15, wherein the graph drawing unit ranks the divergence degree of the change by a plurality of threshold values, and draws a graph showing the change with a different color scheme for each rank.   The graph drawing unit draws a graph indicating a maximum value, a minimum value, and an average value of the reference data and a graph indicating a maximum value, a minimum value, and an average value of the analysis data with different colors. The abnormality cause estimation device according to 15 or 16.   The abnormality cause estimation device according to claim 15, wherein the graph drawing unit further draws a line corresponding to the normalized upper limit value, the standard center value, and the lower limit value. All process tracking data including process management data values for the target product type and production lot from the input device, and standard upper and lower limit values that define the range of the process management data values that satisfy the standard, and the standard value thereof A process tracking all data acquisition process for acquiring standard value data including at least one of the center value,
Based on the standard deviation of the process management data value of a predetermined inspection item from all the process tracking data, the process management data value is selectively acquired and set as reference data, and based on the reference data and the standard value data A process tracking reference data setting step for setting process tracking reference data composed of values obtained by normalizing the process management data values set in the reference data for each process management item;
From all the process tracking data, based on at least one of a predetermined product type and production lot, the process management data value is selectively acquired and set as analysis data, and based on the analysis data and the standard value data, For each process management item, a process tracking analysis data setting process for setting process tracking analysis data consisting of values obtained by normalizing the process management data values included in the analysis data;
For each process management item, a fluctuation calculation step for calculating fluctuations in the analysis data with respect to the reference data,
A graph drawing step of drawing a graph indicating the fluctuation on a display device;
An abnormality cause estimation method comprising: On the computer,
All process tracking data including process management data values for the target product type and production lot from the input device, and standard upper and lower limit values that define the range of the process management data values that satisfy the standard, and the standard value thereof A process tracking all data acquisition process for acquiring standard value data including at least one of the center value,
Based on the standard deviation of the process management data value of a predetermined inspection item from all the process tracking data, the process management data value is selectively acquired and set as reference data, and based on the reference data and the standard value data A process tracking reference data setting step for setting process tracking reference data composed of values obtained by normalizing the process management data values set in the reference data for each process management item;
From all the process tracking data, based on at least one of a predetermined product type and production lot, the process management data value is selectively acquired and set as analysis data, and based on the analysis data and the standard value data, For each process management item, a process tracking analysis data setting process for setting process tracking analysis data consisting of values obtained by normalizing the process management data values included in the analysis data;
For each process management item, a fluctuation calculation step for calculating fluctuations in the analysis data with respect to the reference data,
A graph drawing step of drawing a graph indicating the fluctuation on a display device;
An abnormal cause estimation program that executes

Images