JP2020197914A - Apparatus, method and program for calculating manufacturing condition - Google Patents

Abstract

To provide an apparatus for calculating a manufacturing condition capable of quantitatively carrying out a tackling determination for apparatus abnormality.SOLUTION: An apparatus for calculating a manufacturing condition divides actual measurement data every unit analysis-time period (S2), calculates a histogram feature quantity of a measurement result every combination of measurement items and apparatuses and unit analysis-time periods (S3), calculates a Mahalanobis distance every combination of measurement items and apparatuses and unit analysis-time periods based on the histogram feature quantity (S4), calculates an aggregate Mahalanobis distance based on an aggregate value of Mahalanobis distances over the combinations of measurement items and apparatuses in the latest object-of-analysis period (S5), calculates a process capability index every combination of measurement items and apparatuses and unit analysis-time periods (S6), calculates an aggregate process capability index based on the aggregate value of process capability indexes over the combinations of measurement items and apparatuses in the latest object-of-analysis period (S7), and displays the aggregate Mahalanobis distance and the aggregate process capability in a separate/divisional fashion (S8).SELECTED DRAWING: Figure 3

Description

The present invention relates to anomaly detection technology for chemical plants, steelmaking plants, semiconductor manufacturing equipment, general industrial product manufacturing equipment, etc., and in particular, determines the abnormality detection accuracy of plants and manufacturing equipment based on the operation results during a plurality of past operations. It relates to a manufacturing condition calculation device, a manufacturing condition calculation method, and a manufacturing condition calculation program for enhancing.

In a chemical plant, the same product is often produced using a plurality of devices (reactor, etc.). In this case, it is necessary to appropriately manage the operating state of each device so that the products produced are uniform.
However, when multiple devices are operated in parallel under the same conditions, even if the catalog specifications of each device are the same, there are actually individual differences in the operation results, and the operation results differ for each machining head in the device. In addition, the operating results may differ even if they are produced under the same conditions due to the difference in changes over time of the equipment, and the operating results may change significantly due to the influence of the previous process. Due to such device differences, processing head differences, changes over time, and the influence of the previous process, it is difficult to accurately manage the state in which the produced products are uniform for each device.

Similar problems occur not only in chemical plants but also in semiconductor manufacturing facilities, metal refining plants, water treatment facilities, gas plants and the like. Moreover, such a problem can occur not only in a facility that manufactures something but also in a facility for changing the characteristics of an intangible object such as a power generation / substation facility. Hereinafter, in the present specification, not only a facility for manufacturing something but also a facility for changing the characteristics of an intangible object is referred to as a "plant".
Further, not only complex facilities but also single facilities that behave differently depending on operating conditions are included in the "plant" here.

In the conventional plant control information generator, the cause of an abnormality in the plant is estimated by using the relationship between the past samples.
More specifically, in Patent Document 1, as one of the 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, an abnormality is obtained by pattern recognition using Mahalanobis distance. A method of diagnosing the cause is disclosed.
FIG. 10 shows a flow chart of the abnormality cause diagnosis method of Patent Document 1.

As shown in FIG. 10, first, an appropriate amount of predetermined data (for example, temperature, pressure, etc.) in the steady state of the machine to be diagnosed is measured and used as reference data, and the feature quantities of these reference data are used. Ask. Then, the features are standardized, a correlation matrix is obtained based on the correlation of all combinations between the standardized features, and then the inverse matrix (that is, Mahalanobis space) of the correlation matrix is obtained (step 1). Next, the Mahalanobis distance (MD) of the machine to be diagnosed is determined (step 2). Next, the preset threshold value (generally 0 to 3) is compared with the MD (step 3), and when the MD is larger than the threshold value, the maximum value of the distance element value is specified (step 4). Then, the feature amount corresponding to the maximum value of the distance element value is replaced with the average value of the feature amount of the reference data (step 5). Then, MD is obtained again using the replaced feature amount (step 2). Next, the newly obtained MD is compared with the threshold value (step 3), and if the MD is larger than the threshold value, the maximum value of the distance element value is specified (step 4). Then, in the same manner as described above, the replacement process is performed (step 5), and the process is repeated until the MD becomes smaller than the threshold value. Then, when the MD becomes smaller than the threshold value, the value when the distance element value becomes maximum is set as the value "final effect" indicating the cause of the abnormality of the feature amount as the aggregation of the distance element values.

Japanese Unexamined Patent Publication No. 2004-227279

However, in the conventional configuration of Patent Document 1, since it is necessary to process the data of individual samples as a processing flow, there is a problem that it takes a long time to calculate when the number of samples becomes huge. In addition, the magnitude relationship (direction of abnormality) and the degree of deviation (degree of abnormality) between the process control data estimated as the cause of the abnormality and the process control data in the normal state are unknown, and therefore information on the direction and degree of abnormality. There was a problem that it was difficult to formulate an accurate countermeasure plan for abnormalities based on the diagnosis.
In particular, in general industrial product manufacturing equipment where the parameters that determine the manufacturing process are not comprehensively measured, for example, non-measurement items in the previous process may have a dominant effect on the process, in which case it is abnormal. However, there is a problem that the utilization of abnormality detection becomes insufficient because it may be detected that the operating state of the device has changed in a better direction than before.

An object of the present invention is to solve the above-mentioned problems, and quantitatively determine an approach to an equipment abnormality even in general processing equipment, for example, in which parameters that determine a manufacturing process are not comprehensively measured. An object of the present invention is to provide a manufacturing condition calculation device, a manufacturing condition calculation method, and an abnormality detection program that can be used.

In order to achieve the above object, the manufacturing condition calculation device of the present invention divides the actual measurement data relating to the manufacture of the product for each unit analysis period, and for each combination of the measurement items and devices included in the actual measurement data and the unit analysis period. In addition, a Mahalanobis distance for each combination of the measurement item, the device, and the unit analysis period based on the histogram feature amount calculation unit that calculates the histogram feature amount of the measurement result included in the actual measurement data and the histogram feature amount. The Mahalanobis distance calculation unit that calculates the total Mahalanobis distance based on the total value of the Maharanobis distance for each combination of the measurement item and the device in the latest analysis target period, and the total Mahalanobis distance calculation unit that calculates the total Maharanobis distance, and the measurement item. Aggregate based on the process capability index calculation unit that calculates the process capability index for each combination of the device and the unit analysis period, and the aggregated value of the process capability index for each combination of the measurement items and the device in the latest analysis target period. It has a total process capability index calculation unit that calculates a process capability index, and an abnormality detection display unit that divides and displays the total Mahalanobis distance and the total process capability index.

In the manufacturing condition calculation method of the present invention, the actual measurement data relating to the manufacture of a product is divided for each unit analysis period, and each combination of the measurement items and devices included in the actual measurement data and the unit analysis period is included in the actual measurement data. A histogram feature calculation step for calculating the histogram feature of the measurement result, and a Mahalanobis distance calculation step for calculating the Mahalanobis distance for each combination of the measurement item, the device, and the unit analysis period based on the histogram feature. , The total Mahalanobis distance calculation process for calculating the total Mahalanobis distance based on the total value of the Maharanobis distance for each combination of the measurement item and the device in the latest analysis target period, and the measurement item, the device, and the unit analysis period. A process capability index calculation process that calculates a process capability index for each combination, and a total process capability index that calculates a total process capability index based on the total value of the process capability index for each combination of the measurement items and the equipment in the latest analysis target period. It has a capability index calculation step, and an abnormality detection display step of dividing and displaying the aggregated Mahalanobis distance and the aggregated process capability index.

The manufacturing condition calculation program of the present invention divides the actual measurement data relating to the manufacture of a product into a computer for each unit analysis period, and the actual measurement data for each combination of the measurement items and devices included in the actual measurement data and the unit analysis period. Maharanobis distance that calculates the Maharanobis distance for each combination of the measurement item, the device, and the unit analysis period based on the histogram feature amount calculation process that calculates the histogram feature amount of the measurement results included in The calculation process, the total Maharanobis distance calculation process for calculating the total Maharanobis distance based on the total value of the Maharanobis distance for each combination of the measurement item and the device in the latest analysis target period, and the measurement item, the device, and the unit. Calculate the total process capability index based on the process capability index calculation process that calculates the process capability index for each combination of analysis periods and the total value of the process capability index for each combination of the measurement items and the equipment in the latest analysis target period. The aggregation process capability index calculation step and the abnormality detection display step of dividing and displaying the aggregation Maharanobis distance and the aggregation process capability index are executed.

As described above, according to the manufacturing condition calculation device, the manufacturing condition calculation method, and the manufacturing condition calculation program of the present invention, it is possible to quantitatively carry out the determination of the approach to the device abnormality.

Block diagram of the manufacturing condition calculation device according to the first and second embodiments of the present invention. Block diagram of the processing unit of the manufacturing condition calculation device according to the first and second embodiments. A flowchart showing the processing of the manufacturing condition calculation device according to the first embodiment. The figure which shows the example of the data registered in the database which concerns on Embodiment 1. The figure which shows the example which shows the degree of change and the degree of soundness which concerns on Embodiment 1 at the same time. A flowchart showing the processing of the manufacturing condition calculation device according to the second embodiment. A flowchart showing a part of the details of the processing of the manufacturing condition calculation device according to the second embodiment. The figure which shows the example of the data registered in the database which concerns on Embodiment 2. The figure which shows the example which shows the degree of change and the degree of soundness which concerns on Embodiment 2 at the same time. Flow chart showing the conventional method of diagnosing the cause of an abnormality

(Embodiment 1)
FIG. 1 is a block diagram of the manufacturing condition calculation device 1 according to the first embodiment of the present invention. Figure 2
Is a block diagram of the processing unit 166 of the manufacturing condition calculation device 1. The manufacturing condition calculation device 1 according to the first embodiment can be realized by computer system hardware and a program executed on the computer system. The manufacturing condition calculation device 1 shown here is merely an example, and other configurations can be used.

<Device>
As shown in FIG. 1, the manufacturing condition calculation device 1 includes a computer 120, a monitor 122, a printer 124, a keyboard 126, and a mouse 128, all connected to the computer 120.

The computer 120 includes a drive 150 of a DVD-ROM (Digital Versaille Disk Read-Only-Memory) and a semiconductor memory port 152. Further, the computer 120 includes a bus 142 connected to the DVD-ROM drive 150 and the semiconductor memory port 152, a CPU 140 all connected to the bus 142, and a ROM 144 for storing the boot-up program of the computer 120. The computer 120 also includes a RAM 146, a hard disk drive 148, and a network interface 154. The RAM 146 provides a work area used by the CPU 140 and is a storage area for a program executed by the CPU 140. The hard disk drive 148 stores voice data, an acoustic model, a language model, a lexicon, and a mapping table. Network interface 154 provides a connection to network 164.

With reference to FIG. 2, the manufacturing condition calculation device 1 includes a processing unit 166 configured by the CPU 140 executing a program. The processing unit 166 includes an actual measurement data recording unit 168, an analysis period setting unit 170, a histogram feature amount calculation unit 172, a Mahalanobis distance calculation unit 174, a total Mahalanobis distance calculation unit 176, a process capability index calculation unit 178, and the like. It includes a total process capability index calculation unit 180 and an abnormality detection display unit 182.

The software that realizes the processing unit 166 of the manufacturing condition calculation device 1 according to the first embodiment is distributed in the form of an object code or a script recorded on a medium such as a DVD-ROM 162 or a semiconductor memory 160, and is a DVD-ROM drive. It is provided to the computer 120 via a reading device such as 150 or the semiconductor memory port 152 and stored in the hard disk drive 148. When the CPU 140 executes the program, the program is read from the hard disk drive 148 and loaded into the RAM 146. Then, the instruction is fetched from the address specified by the program counter (not shown), and the instruction is executed. Further, the CPU 140 reads data to be processed from the hard disk drive 148, and stores the processing result in the hard disk drive 148 as well. A hard copy of the estimated product molecular weight is printed by printer 124.
Since the general operation of the computer 120 is well known, detailed description thereof will be omitted.

Regarding the distribution method of software, the software does not necessarily have to be recorded on a storage medium. For example, the software may be distributed from another computer connected to the network. Further, a part of the software may be stored in the hard disk drive 148, and the remaining part of the software may be taken into the hard disk drive 148 via the network and integrated at the time of execution.
Moreover, the distribution form of software is not limited to object code. As described above, it may be in script format, or it may be distributed in the form of a source program and converted into object code by an appropriate compiler installed on the computer 120.

Typically, modern computers utilize the general functionality provided by an operating system (OS) to achieve functionality in a controlled manner according to the desired purpose. Therefore, even if the program does not include general functions that can be provided by the OS or a third party and only specifies a combination of execution orders of general functions, the program as a whole achieves a desired purpose. It is clear that the program is included in the scope of the present invention as long as it has.

<Flow>
Next, a flowchart of the data analysis process in the manufacturing condition calculation device 1 will be described with reference to FIG.

<S1>
First, in step S1, the actual measurement data recording unit 168 of the processing unit 166 adds the actual measurement data to the database constructed in the computer 120 (hereinafter, simply referred to as “DB”). The measured data when registered for each individual to be manufactured includes the variety and serial number of the individual to be manufactured, the manufacturing time in each manufacturing process, the manufacturing equipment, the manufacturing head, and the manufacturing result.
The manufacturing result is a manufacturing result in each manufacturing process, for example, a product width or a product length.

<S2>
Next, in step S2, the analysis period setting unit 170 sets the unit analysis period to, for example, 6 hours based on the analysis target, data, and evaluation candidates. Here, there is a production lot according to the unit analysis period. In this lot, by flowing the individual to be manufactured in a certain quantity, for example, a mass of 1,000 or 10,000 pieces, the individual lots are transported from the pre-process to the post-process, and the production data is aggregated. It is used to do things. Since the quantity of this production lot is set so as to easily reflect the actual situation of each manufacturing site, the lot number for managing the production lot will be described as an index of the unit analysis period in the figures of future explanations.

<S3>
Next, in step S3, the histogram feature amount calculation unit 172 divides the measured data for each unit analysis period, calculates the histogram feature amount of the measurement result for each combination of the device, the processing head, the lot number, and the measurement item, and DB. Register with.
Specifically, the histogram feature amount is the average value, median value, standard deviation, mode value, kurtosis, skewness, etc. of the measurement results. By calculating the features based on the combination of median, standard deviation, kurtosis, and skewness, generally good results can often be obtained.

FIG. 4 shows an example of the result of calculating the histogram feature amount for each unit analysis period from the measured data from the process to the average.
In the first embodiment, the histogram feature amount of the measurement result is calculated for each combination of the target process, apparatus, processing head, lot number, and measurement item.

First, the process is the process of processing the individual to be manufactured from the material to the finished form. For example, in the battery manufacturing process, processes such as kneading, coating, winding, can insertion, welding, liquid injection, sealing, initial inspection, aging, and shipping inspection are shown. In the first embodiment, while assuming a manufacturing process including steps 1 to 10, only step 1 is partially displayed, for example, only a winding step.

Next, the device shown in FIG. 4 is a name for a machine that processes an individual to be manufactured. Depending on the process, some processes have only one device, while others have multiple devices. In the first embodiment, the partial results of the devices 1 to 4 are displayed while assuming that there are four devices in the step 1.

The processing head shown in FIG. 4 is the name of the processing head. In order to increase productivity in the equipment, a plurality of parts that perform the same processing may be provided. In the first embodiment, it is assumed that each apparatus in the process 1 has two processing heads, and the partial results of the processing heads 1 to 2 are displayed.

The lot number shown in FIG. 4 is a name that uniquely identifies the lot. This lot means that the individual to be manufactured is flowed in a certain quantity, for example, a mass of 1,000 or 10,000 pieces, so that the individual lots are transported from the pre-process to the post-process, and the production data is aggregated. It is a unit used to do things. The quantity of this production lot is set so that it can easily reflect the actual situation of each manufacturing site. In the first embodiment, the lot number for managing the production lot is used in the description as an index according to the unit analysis period.

The measurement items shown in FIG. 4 are items measured for the purpose of guaranteeing the processing result in the apparatus and items measured for the purpose of recording fluctuations in processing conditions. is there. For example, assuming a winding process, the measurement items are the winding body height, the winding body diameter, the winding deviation amount, the winding tension, the winding correction amount, and the like. In the first embodiment, the partial results of A and D are displayed while assuming that the measurement items are from A to G.

Then, after stratifying the measured data for each combination of these processes, equipment, processing heads, lot numbers according to the unit analysis period, and measurement items, the number of constituents N as the histogram feature amount, the median value of the measurement results, the standard deviation, and so on. The kurtosis, skewness, and average are calculated. The histogram feature amount may not include at least one of the median value, standard deviation, kurtosis, skewness, and average, and calculated values other than these may be calculated as the histogram feature amount.

Here, the skewness is calculated using the following equation (1).

The kurtosis is calculated using the following formula (2).

<S4>
Next, in step S4, the Mahalanobis distance calculation unit 174 calculates the Mahalanobis distance MD for each combination of the device, the processing head, the lot number, and the measurement item, including the histogram feature amount registered in the past by the histogram feature amount calculation unit 172. Calculate and register in DB.
Specifically, among the histogram features, the Mahalanobis distance is calculated based on the data of the past month with the combination of median, standard deviation, kurtosis, and skewness, and generally good results can be obtained. Often.

FIG. 4 shows an example of the result of calculating the Mahalanobis distance MD.
Here, the Mahalanobis distance MD is calculated using the following equation (3), assuming a p-dimension in which there is no correlation between different variables. In the first embodiment, the Mahalanobis distance MD is calculated with i = 1 as the median, i = 2 as the standard deviation, i = 3 as the kurtosis, and i = 4 as the skewness.

<S5>
Next, in step S5, the total Mahalanobis distance calculation unit 176 refers to the total Mahalanobis distance for each combination of the device, the processing head, and the measurement item in the latest analysis target period with respect to the Mahalanobis distance MD calculated by the Mahalanobis distance calculation unit 174. A_MD is calculated and registered in the DB.
Specifically, the latest analysis target period is set to 3 days, and the average value of the Mahalanobis distance MD for each combination of equipment, processing head, and measurement item in this analysis target period is calculated as the aggregated Mahalanobis distance A_MD, which is generally good. Results are often obtained.
FIG. 4 shows an example of the result of calculating the aggregated Mahalanobis distance A_MD using the data of the Mahalanobis distance MD. For example, the Mahalanobis distance MD of the combination of the device 1, the processing head 1, and the measurement item A in the lot numbers of the analysis target period such as AAAAA001, AAAAA004 to AAAA008 is 0.91,2.43, 1.71, 3.72, respectively. , 2.32, 2.14, .... The total Mahalanobis distance calculation unit 176 calculates 2.21 of these average values as the total Mahalanobis distance A_MD.

<S6>
Next, in step S6, the process capability index calculation unit 178 divides the measured data for each unit analysis period, calculates the process capability index for each combination of the device, the processing head, the lot number, and the measurement item, and registers the process capability index in the DB. ..
Specifically, as the process capability index, generally good results can often be obtained by using the process capability index Cpk in consideration of the bias of the average.
FIG. 4 shows an example of the result of calculating the process capability index Cpk using the upper standard value UCL and the lower standard value LCL of the measurement items.
Here, the process capability index Cpk is calculated using the following formula (4), where the upper standard value of each measurement item is UCL and the lower standard value is LCL.

<S7>
Next, in step S7, the total process capability index calculation unit 180 sets the process capability index Cpk calculated by the process capability index calculation unit 178 for each combination of device, processing head, and measurement item in the latest analysis target period. Aggregate process capability index A_Cpk is aggregated and registered in the DB.
Specifically, the latest analysis target period is set to 3 days, and the average value of the process capability index Cpk for each combination of equipment, processing head, and measurement item in this analysis target period is calculated as the aggregate process capability index A_Cpk. In general, good results are often obtained.
FIG. 4 shows an example of the result of calculating the aggregated process capability index A_Cpk using the data of the process capability index Cpk. For example, the process capability indexes Cpk of the combination of the device 1, the processing head 1, and the measurement item A in the lot numbers of the analysis target period such as AAAAA001, AAAAA004 to AAAA008 are 3.41, 3.72, 3.57, and 2. 9, 3, 3.42, ... The aggregation process capability index calculation unit 180 calculates the average value of 3.34 as the aggregation process capability index A_Cpk.

<S8>
Next, in step S8, the abnormality detection display unit 182 combines the total Mahalanobis distance A_MD and the total process capability index A_Cpk and outputs the combination to the monitor 122 or the like, which is confirmed by the operator at the manufacturing site.
FIG. 5 shows an example of the results of graphing the total Mahalanobis distance A_MD representing the degree of change on the horizontal axis and the total process capability index A_Cpk representing the degree of soundness on the vertical axis.
For example, the lower right region of FIG. 5 has a low degree of soundness and a large degree of recent change, indicating that there is a possibility of indicating a sudden abnormality. Therefore, the operator immediately investigates the current situation and considers whether to implement countermeasures.
On the other hand, the lower left region has a low degree of soundness and a small degree of recent change, indicating that it may indicate a chronic abnormality. For this reason, operators need to continuously consider improvement plans, although the priority of their efforts is low.
Furthermore, the upper right region shows that the degree of soundness is high but the degree of recent change is large. Therefore, the operator needs to continuously monitor whether or not a sudden abnormality occurs.
And since the upper left area has a high degree of soundness and a small degree of recent change, the priority of efforts is extremely low.

In the process of going through such a flow, even in general processing equipment where the parameters that determine the manufacturing process are not comprehensively measured, for example, even in general processing equipment, while evaluating changes in the items to be measured as a statistical model, the equipment is in operation. It is possible to display related data that is directly linked to deterioration. Then, by checking the display result from the above viewpoint, the operator can detect changes in material properties that are difficult to measure and record and abnormalities that are not easily affected by the previous process. Judgment can be made quantitatively.
The present invention is not limited to the above embodiment, and can be implemented in various other aspects.

The second embodiment of the present invention will be described. The difference between the first embodiment and the second embodiment is the processing contents of the total Mahalanobis distance calculation unit 176 and the abnormality detection display unit 182.

<Flow>
The flowchart of the data analysis process in the manufacturing condition calculation device 1 will be described with reference to FIGS. 6 and 7. The flow of the second embodiment is the same as that of the first embodiment up to step S7. Here, in the second embodiment, it is not necessary to perform the process of step S5.

<S11>
After the processing of step S7 is performed, as shown in FIG. 6, in step S11, the total Mahalanobis distance calculation unit 176 uses the device / machining head / lot based on the process capability index Cpk and the total process capability index A_Cpk. Calculate the modified Mahalanobis distance MD'for each combination of No. and measurement items. As shown in FIG. 7, the process of step S11 includes the process of steps S21 to S23.
<S21>
In step S21, the total Mahalanobis distance calculation unit 176 determines whether or not the process capability index Cpk exceeds the total process capability index A_Cpk.
<S22>
When the total Mahalanobis distance calculation unit 176 determines that the process capability index Cpk exceeds the total process capability index A_Cpk, in step S22, the value obtained by multiplying the Mahalanobis distance MD by "-1" is calculated as the modified Mahalanobis distance MD'. ..
<S23>
When the total Mahalanobis distance calculation unit 176 determines that the process capability index Cpk does not exceed the total process capability index A_Cpk, in step S23, the value obtained by multiplying the Mahalanobis distance MD by "1" is calculated as the modified Mahalanobis distance MD'. ..

<S12>
Next, after the processing of step S11 is performed, in step S12, the total Mahalanobis distance calculation unit 176 registers the modified Mahalanobis distance MD'in the DB.
FIG. 8 shows an example of the result of calculating the modified Mahalanobis distance MD'using the data of the Mahalanobis distance MD.

<S13>
Next, in step S13, the total Mahalanobis distance calculation unit 176 calculates the total modified Mahalanobis distance A_MD'for each combination of the device, the processing head, and the measurement item in the latest analysis target period with respect to the modified Mahalanobis distance MD'. Register in the DB.
Specifically, the latest analysis target period is set to 3 days, and the average value of the modified Mahalanobis distance MD'for each combination of equipment, processing head, and measurement item in this analysis target period is calculated as the aggregated modified Mahalanobis distance A_MD'. In many cases, good results can be obtained.

FIG. 8 shows an example of the result of calculating the aggregated modified Mahalanobis distance A_MD'using the data of the modified Mahalanobis distance MD'. For example, the modified Mahalanobis distance MD'of the combination of the device 1, the processing head 1, and the measurement item A in the lot numbers of the analysis target period such as AAAAA001, AAAAA004 to AAAA008 is -0.9, -2.4, -1, respectively. 7,3.72,2.32. -2.1, ... The aggregated Mahalanobis distance calculation unit 176 calculates the average value of −0.2 as the aggregated modified Mahalanobis distance A_MD ′.

<S14>
Next, in step S14, the abnormality detection display unit 182 combines the total correction Mahalanobis distance A_MD'and the total process capability index A_Cpk and outputs it to the monitor 122 or the like, and the operator at the manufacturing site confirms it.
FIG. 9 shows an example of the results of graphing the total correction Mahalanobis distance A_MD'representing the degree of change on the horizontal axis and the total process capability index A_Cpk representing the degree of soundness on the vertical axis.
For example, the lower right region of FIG. 9 shows that the soundness is low and the degree of recent change is large in the positive direction, which may indicate a sudden deterioration abnormality. Therefore, the operator needs to investigate the current situation immediately and take measures.
On the other hand, the lower center region has a low degree of soundness and a small degree of recent change, indicating that it may indicate a chronic abnormality. For this reason, operators need to continuously consider improvement plans, although the priority of their efforts is low.
Furthermore, the lower left region has a low degree of soundness and a large degree of recent change in the negative direction, which may indicate a sudden improvement abnormality. For this reason, operators need to confirm what kind of changes have occurred, although the priority of their efforts is low, and stock them as future improvement plans.
In addition, since the upper region has a high degree of soundness as a whole, the operator continuously monitors whether the degree of soundness deteriorates.

By confirming the display result from such a viewpoint, the operator can quantitatively make a decision on the approach to the device abnormality.
The present invention is not limited to the above embodiment, and can be implemented in various other aspects.

The manufacturing condition calculation device, the manufacturing condition calculation method, and the manufacturing condition calculation program according to the present invention are difficult to measure and record even in, for example, general processing equipment and lines where the parameters that determine the manufacturing process are not comprehensively measured. It will be possible to quantitatively evaluate the degree of change due to changes in material properties and equipment fluctuations, making it possible to quantitatively make decisions on changes in manufacturing conditions, and accelerating the use of conventional acquired parameters in general factories, etc. It is useful for things such as.

1 Manufacturing condition calculation device 120 Computer 122 Monitor 124 Printer 126 Keyboard 128 Mouse 140 CPU
142 bus 144 ROM
146 RAM
148 Hard Disk Drive 150 DVD-ROM Drive 152 Semiconductor Memory Port 154 Network Interface 160 Semiconductor Memory 162 DVD-ROM
164 Network 166 Processing unit 168 Actual measurement data recording unit 170 Analysis period setting unit 172 Histogram feature amount calculation unit 174 Mahalanobis distance calculation unit 176 Total Mahalanobis distance calculation unit 178 Process capability index calculation unit 180 Total process capability index calculation unit 182 Abnormality detection display unit

Claims (7)

The actual measurement data related to the manufacture of the product is divided for each unit analysis period, and the histogram feature amount of the measurement result included in the actual measurement data is calculated for each combination of the measurement items and devices included in the actual measurement data and the unit analysis period. Histogram feature calculation unit and
A Mahalanobis distance calculation unit that calculates the Mahalanobis distance for each combination of the measurement item, the device, and the unit analysis period based on the histogram feature amount.
The total Mahalanobis distance calculation unit that calculates the total Mahalanobis distance based on the total value of the total Mahalanobis distance for each combination of the measurement item and the device in the latest analysis target period,
A process capability index calculation unit that calculates a process capability index for each combination of the measurement item, the device, and the unit analysis period.
A total process capability index calculation unit that calculates a total process capability index based on the total value of the process capability index for each combination of the measurement item and the device in the latest analysis target period.
A manufacturing condition calculation device having an abnormality detection display unit that divides and displays the total Mahalanobis distance and the total process capability index. The manufacturing condition calculation device according to claim 1, wherein the histogram feature amount calculation unit calculates at least one of the median value, standard deviation, kurtosis, and skewness of the measurement result as the histogram feature amount. .. The manufacturing condition calculation device according to claim 1 or 2, wherein the total Mahalanobis distance calculation unit calculates an average value of the Mahalanobis distances for each combination as the total Mahalanobis distances. The manufacturing condition calculation device according to any one of claims 1 to 3, wherein the total process capability index calculation unit calculates an average value of the process capability indexes for each combination as the total process capability index. The total Mahalanobis distance calculation unit compares the process capability index with the total process capability index for each combination of the measurement item, the device, and the unit analysis period, and when the process capability index exceeds the total process capability index. , A negative code is given to the Mahalanobis distance corresponding to the process capability index, and a positive code is given to the Mahalanobis distance corresponding to the process capability index when the process capability index does not exceed the total process capability index. The manufacturing condition calculation device according to any one of claims 1 to 4, wherein the total Mahalanobis distance is calculated based on the Mahalanobis distance to which a positive and negative reference numerals are given. The actual measurement data related to the manufacture of the product is divided for each unit analysis period, and the histogram feature amount of the measurement result included in the actual measurement data is calculated for each combination of the measurement items and devices included in the actual measurement data and the unit analysis period. Histogram feature calculation process and
A Mahalanobis distance calculation step for calculating a Mahalanobis distance for each combination of the measurement item, the device, and the unit analysis period based on the histogram feature amount.
The total Mahalanobis distance calculation process for calculating the total Mahalanobis distance based on the total value of the total Mahalanobis distance for each combination of the measurement item and the device in the latest analysis target period.
A process capability index calculation process that calculates a process capability index for each combination of the measurement item, the device, and the unit analysis period, and
A total process capability index calculation process that calculates a total process capability index based on the total value of the process capability index for each combination of the measurement item and the device in the latest analysis target period.
A manufacturing condition calculation method including an abnormality detection display step of dividing and displaying the total Mahalanobis distance and the total process capability index. On the computer
The actual measurement data related to the manufacture of the product is divided for each unit analysis period, and the histogram feature amount of the measurement result included in the actual measurement data is calculated for each combination of the measurement items and devices included in the actual measurement data and the unit analysis period. Histogram feature calculation process and
A Mahalanobis distance calculation step for calculating a Mahalanobis distance for each combination of the measurement item, the device, and the unit analysis period based on the histogram feature amount.
The total Mahalanobis distance calculation process for calculating the total Mahalanobis distance based on the total value of the total Mahalanobis distance for each combination of the measurement item and the device in the latest analysis target period.
A process capability index calculation process that calculates a process capability index for each combination of the measurement item, the device, and the unit analysis period, and
A total process capability index calculation process that calculates a total process capability index based on the total value of the process capability index for each combination of the measurement item and the device in the latest analysis target period.
A manufacturing condition calculation program that executes an abnormality detection display process that divides and displays the total Mahalanobis distance and the total process capability index.

Images