JP2001273336A - Data collection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To increase collection frequency of the data to be managed and to shorten a period during which nonstandard data are generated when the data to be managed show the abnormal variance and may possibly get out of the standard. SOLUTION: This data collection method includes a data collection means 1 which measures the data to be managed and stores them in a data base 30 that collects the measured value, a data analysis means 2 which decides whether the data collected by the means 1 show a specific variance by means of the management method using the management value registered in a data base 40 and a collection frequency control means 30 which changes the data collection frequency of the means 1 into the new collection frequency registered in a data base 50 when the specific variance of the data is decided by the means 2. In such a constitution, the data to be managed can be measured and checked more frequently than before during the period that the data are showing the abnormal variance.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a method for collecting data to be managed (managed data) which is periodically measured and collected in order to manage product quality and performance of a manufacturing apparatus. It is about.

[0002]

2. Description of the Related Art As a conventional data collection technique, data to be managed is always measured and collected at regular intervals, regardless of how the data to be managed changes. Moreover, the collection of the management target data is not always performed in real time, and a certain number of days may elapse before the next data collection, such as every other week or every third day.

[0003]

However, according to the conventional data collection method, even when there is a sign that the managed data is likely to deviate from the standard, the managed data is collected in the same cycle as the time when the data showed a stable transition. Measuring and managing product quality and manufacturing equipment. For this reason, there has been a situation where, for a long time until the next measurement date, the data to be managed does not notice that it is out of the standard.

Accordingly, an object of the present invention is not to measure the data in real time, but to leave the data to be managed in an abnormal state with respect to the data to be measured, which is measured periodically, and to leave the data out of the standard. The purpose of the present invention is to provide a data collection method for preventing data from being collected.

[0005]

According to a first aspect of the present invention, there is provided a data collection method comprising: a data collection step of measuring and collecting data to be managed; and a data collection method in the data collection step. However, a data analysis step of determining whether or not a specific change has been shown by a management method using a management value, and when it is determined that the data has shown a specific change in the data analysis step, data collection in the data collection step A collection frequency adjusting step of changing the frequency.

[0006] As described above, the data collected in the data collection step determines whether or not the data shows a specific change by the management method using the control value. If it is determined that the data collection has been performed, the data analysis process includes a collection frequency adjustment process for changing the data collection frequency in the data collection process. When an abnormal change is detected, the collection frequency adjustment step instructs the data collection step to change the collection frequency of the management target data to a higher collection frequency. As a result, while the managed data shows abnormal fluctuations,
Measurement and check of the management target data can be performed more frequently than before, and it is possible to prevent the management target data from being abnormally changed and left as it is out of the standard.

According to a second aspect of the present invention, there is provided a data collection method, comprising: a data collection step of measuring and collecting data to be managed; and whether the data collected in the data collection step shows a specific change by a management method using a management value. A data analysis step of determining whether or not the data collection step includes a collection frequency adjustment step of changing the data collection frequency of the data collection step, and whether the data analyzed in the data analysis step is within a standard range. A collection frequency evaluation step of evaluating whether or not the collection frequency is an appropriate value, and correcting the collection frequency of the data collection step by the collection frequency adjustment step if inappropriate.

As described above, it is evaluated whether or not the collection frequency is an appropriate value based on whether or not the data analyzed in the data analysis step is within the range of the standard. Since a collection frequency evaluation step of correcting the collection frequency by the collection frequency adjustment step is included, it is evaluated whether or not the current collection frequency is appropriate. If the current collection frequency is inappropriate, the collection frequency is corrected so as to increase the collection frequency. By providing this collection frequency evaluation step, even if a person does not notice that the currently set collection frequency has become inappropriate for some reason, the collection frequency is automatically corrected. be able to.

According to a third aspect of the present invention, in the data collection method according to the second aspect, in the collection frequency evaluation step, at least the data shows a specific change in order to evaluate whether or not the collection frequency is an appropriate value. The number of times that data deviates from the standard from the time to the next data collection is counted, and if this count satisfies a predetermined condition, it is determined that the collection frequency is not an appropriate value.

As described above, in the collection frequency evaluation step, in order to evaluate whether or not the collection frequency is an appropriate value, the data is collected at least from the time when the data shows a specific change until the next data collection. Count the number of deviations from the standard, and if this count satisfies the predetermined condition, it is determined that the collection frequency is not an appropriate value, so if the measured value of the data to be managed exceeds the range of the management value, Increase the value of the number of times that the current collection frequency was determined to be inappropriate by one, otherwise increase the value of the number of times that the current collection frequency was determined to be valid by one, and The validity of the collection frequency is evaluated based on whether or not the rate at which the current collection frequency is determined to be inappropriate is not less than a certain value. If the collection frequency is inappropriate, the collection frequency can be corrected.

According to a fourth aspect of the present invention, in the data collection method, a data collection step of measuring and collecting data to be managed, and whether the data collected in the data collection step shows a specific change by a management method using a management value. A data analysis step of determining whether or not the data has undergone a specific change in the data analysis step, and a collection frequency change notification step of notifying the user to change the data collection frequency of the data collection step. .

Thus, instead of the collection frequency adjusting step of the data collection method according to the first aspect, if it is determined that the data shows a specific change in the data analysis step, the data collection frequency in the data collection step is changed. In such a case, the data to be managed cannot be automatically collected by the measuring device, and even if the data to be managed is manually sent to the data collection process, When an abnormal change is indicated, it is possible to instruct a person to increase the collection frequency of the management target data. As a result, the frequency of data collection in the data collection process is increased.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a functional block diagram according to a first embodiment of the present invention, FIG. 2 is an explanatory diagram showing an example of a management method for detecting abnormal fluctuation of management target data, and FIG. 3 is a table structure of a database used in a data collection process. , FIG. 4 is an explanatory view of a table structure of a database used in a data analysis step, FIG. 5 is an explanatory view of a table structure of a database used in a collection frequency adjustment step, and FIG. 6 is a diagram showing a change of a collection frequency in the first embodiment. It is explanatory drawing which shows an example of.

This data collection method controls a device for measuring data to be managed, and performs a data collection step of measuring and collecting data. ), And a collection frequency adjustment step for instructing the data collection step to change the data collection frequency when the data to be managed shows abnormal fluctuations. including. An apparatus provided with means for performing these steps is shown in FIG.

In FIG. 1, reference numeral 1 denotes a data collection unit, 2 denotes a data analysis unit, and 3 denotes a collection frequency adjustment unit. Reference numeral 20 denotes a device for measuring data to be managed (managed data). The data to be managed is sent from the measuring device 20 to the data collection means 1. Reference numerals 30, 40 and 50 are databases.
The database 30 stores management target data used by the data analysis means 2 as time-series data.
The database 40 stores management values used to determine whether or not the management target data indicates an abnormal change. The database 50 stores information on the collection frequency of the management target data.

The data collecting means 1 measures data to be managed using the measuring device 20 and collects the data. The data is stored in the database 30. For example, in a semiconductor manufacturing line, in order to manage the performance of a semiconductor manufacturing apparatus, the number of particles, an etch rate, and the like are periodically measured and stored in a database.

The data analysis means 2 uses the management data stored in the database 30 and the management values stored in the database 40 to determine whether or not the management data indicates an abnormal change. The abnormal fluctuation of the management target data is a fluctuation indicating a sign that the quality of the product or the performance of the manufacturing apparatus is reduced, and as a method of determining whether such a fluctuation has occurred, The QC method or the like is used.

As an example, FIG. 2 shows a method of judging abnormal fluctuation. In FIG. 2, in the case of (a), an upper limit alert value (upper limit value −δ) or a lower limit alert value (lower limit value + δ) is set as a management value stricter than the upper limit value or the lower limit value, and the upper limit alert value or the lower limit value is set. A change in which there is a measurement point exceeding the alert value is determined to be an abnormal change. In the case of (b), a change in which there are N consecutive measurement points whose change amount is D or more is determined to be an abnormal change. In the case of (c), a change in which a measurement point having a change amount equal to or greater than D or a measurement point having a change amount K times the previous change amount is determined to be an abnormal change. In the case of (d), a change in which m out of n consecutive measurement points exist within the range δ from the upper limit value or the lower limit value is determined to be an abnormal change.

The database 30 has a table shown in FIG. 3 for treating the data to be managed as time-series data, and has a dedicated table for each data to be managed. In FIG. 3, f31 is a field for storing the measurement date and time of the data to be managed, and f32 is a field for storing the measurement value.

The database 40 stores management values (upper limit, upper limit alert value,...,..., N consecutive values, and change amount D required for the management method (for example, (a) to (d)) of each data to be managed.
Etc.) are stored. FIG. 4 shows a table structure of the database 40. In FIG. 4, f41 is a field for storing the name of the data to be managed (code name indicating the name of the data to be managed), and f42 is a field for storing the name of the management method (code name indicating the name of the management method). The field f43 is a field for storing the name of the management value (code name indicating the name of the management value), and the field f44 is a field for storing the management value.

For example, in FIG. 4, in the first row of the table, the data to be managed is the number of particles, the management method is (a) in FIG. 2, the name of the management value is the upper limit, and the management value is Represents 10.

The collection frequency adjusting means 3 instructs the data collection means 1 to increase the collection frequency of the data to be managed from the next time when an abnormal change of the data to be managed is detected. The degree to which the collection frequency is increased depends on the database 50.
And is set in advance for each management method of each data to be managed.

When no abnormal change in the data to be managed is detected, the information on the corresponding collection frequency is sent to the data collection means 1 based on the information stored in the database 50.

Here, the table structure of the database 50 is shown in FIG. f51 is a field for storing the name of the data to be managed, f52 is a field for storing the name of the management method, f53 is a field for storing the name of the management value, and f54 is a field for storing the collection frequency. is there.

For example, in FIG. 5, in the first row of the table, the data to be managed is the number of particles, the management method is (a) in FIG. 2, the name of the management value is the upper limit alert value, The frequency value indicates that data is measured and collected every 24 hours. In the case of the second line, the name of the management value is not stored in the field f53, and the collection frequency (data is collected every 72 hours) when the number of particles does not indicate abnormal fluctuation is shown.

FIG. 6 shows an example when the collection frequency changes. In FIG. 6, when the measured value of the data to be managed is less than the upper limit alert value A, the data to be managed is measured every three days.
Collection frequency is set to 3 so that data to be managed is measured every other day
If it exceeds the upper limit alert value B,
The data to be managed is monitored by changing the collection frequency six times so that the data to be managed is measured every half day.

As described above, in the first embodiment of the present invention, by providing the collection frequency adjusting means 3, it becomes possible to adjust the collection frequency of the data to be managed by the data collecting means 1, and Starts to show abnormal fluctuations indicating signs of quality deterioration, the frequency of collection of the managed data can be automatically increased so that the fluctuations of the managed data can be grasped in detail.

FIGS. 7 to 9 show a second embodiment of the present invention.
It will be described based on. FIG. 7 is a functional block diagram according to the second embodiment of the present invention, FIG. 8 is an explanatory diagram of a table structure of a database used in a collection frequency evaluation step, and FIG.
It is an algorithm used in the collection frequency evaluation means in the embodiment.

As shown in FIG. 7, in the functional block diagram of the first embodiment (FIG. 1), a collection frequency evaluation means (collection frequency evaluation step) 4 and a database storing data used by the collection frequency evaluation means 4 60 is added.

The collection frequency evaluation means 4 includes a data analysis means 2
It is evaluated whether the collection frequency is an appropriate value based on whether or not the data analyzed in the above is within the range of the standard. If inappropriate, the collection frequency of the data collection unit 1 is adjusted to the collection frequency adjustment unit 3.
To correct. In this case, whether or not the current collection frequency is appropriate is evaluated, and the evaluation result is recorded in the database 60. The inappropriate collection frequency is the collection frequency when the management target data is out of the standard from the time when the abnormal change of the management target data is detected until the next data collection.

Further, if the current collection frequency is too inappropriate, the database 50 is accessed to correct the value of the collection frequency. The collection frequency evaluation unit 4 is executed before the collection frequency adjustment unit 3 to correct the contents of the database 50.

The database 60 stores information for determining whether or not the current collection frequency is too inappropriate. Here, the table structure of the database 60 is shown in FIG. In FIG. 8, f61 is a field for storing the name of the data to be managed, f62 is a field for storing the name of the management method, and f63 is a field for storing the number of times that the current collection frequency is determined to be appropriate. F64 is a field for storing the number of times that the current collection frequency is determined to be inappropriate, and f65 is a field for storing a condition value or the like for correcting the collection frequency. The data of f61 and the data of f51 are the same, and the data of f62 are the same as the data of f52.

Next, how to determine that the current collection frequency is too inappropriate will be described with reference to the algorithm of FIG. Here, an example relating to the management method of FIG. 2A is used.

In step 1, the measured values of the data to be managed collected from the measuring device are collected from the data collecting means 1. In step 2, a management value for determining whether or not the management target data indicates an abnormal change is stored in the database 4.
Extract from 0. In step 3, if the measured value of the data to be managed exceeds the range from the lower limit to the upper limit, the value of the field f64 of the database 60 is increased by one. Otherwise, the value of the field f63 of the database 60 is increased by one. In step 4, if the sum of the fields f63 and f64 of the database 60 is 5 or more and the ratio of the current collection frequency determined to be inappropriate is 20% or more, the data collection interval is reduced to half. (Field f5 of database 50
4 is halved). * F62, f63 described above
Is considered to correspond to f63 and f64 in Fig. 9, so please modify either one. In FIG. 8,
There are two f64, but please confirm whether this is enough. As described above, in step 4, the validity of the current collection frequency is evaluated, and if the collection frequency is inappropriate, it is corrected.

As described above, according to the second embodiment of the present invention, by providing the collection frequency evaluation means 4 for correcting the collection frequency when the data to be managed shows an abnormal fluctuation, the appropriate collection frequency can be obtained. Can be changed automatically. In addition, the information stored in the database 60 is very useful when examining whether or not the collection frequency of management target data that is normally performed is appropriate.

The collection frequency evaluation means 4 of this embodiment
Corrects the collection frequency when the data to be managed shows an abnormal change, but may be applied to the correction of the collection frequency when the data does not show an abnormal change. In this case, for example, if the change amount is used as a standard value and a situation that shows a change equal to or more than a certain change amount frequently occurs before the next data collection, the collection frequency is corrected. In this way, it is less likely that the user will not notice that there has been a large change in the management target data until the next data collection.

A third embodiment of the present invention will be described with reference to FIGS. FIG. 10 shows a third embodiment of the present invention.
FIG. 11 is an explanatory diagram showing a table structure of a database used in a collection frequency change notification step.

As shown in FIG. 10, in the functional block diagram of the first embodiment (FIG. 1), a collection frequency change notifying means (collection frequency change notification step) 5 and a notification destination are replaced by the collection frequency adjustment means 3. The recorded database 70 is added.

When the data analysis means 2 determines that the data shows a specific change, the collection frequency change notification means 5 notifies the data collection means 1 to change the data collection frequency. In this case, when it is determined that the management target data shows an abnormal change, an instruction is given to a notification destination designated in advance to increase the collection frequency of the management target data. E-mail, a mobile phone, or the like is used as a means for contacting the notification destination.

For example, in the case of an electronic mail, this embodiment has a database 70 in which a mail address of a report destination is recorded, and transmits a message relating to a change in collection frequency using this information and a mail command of a computer. FIG. 11 shows a table structure of the database 70. f71 is a field for storing the name of the data to be managed, f72 is a field for storing an e-mail address, and f73 is a field for storing a message to be sent to the notification destination.

This embodiment having the above-mentioned collection frequency change notifying means 5 is used when manual work is required for collecting data to be managed, for example, for measuring the number of particles attached to a wafer. As a result, even when it is necessary to measure the data to be managed and to transmit the data from the measuring device to the data collecting means 1 by a human, the frequency of collecting the data to be managed is improved as in the first embodiment, and the same as in the first embodiment. The effect can be obtained.

[0042]

According to the data collection method of the first aspect of the present invention, a data analysis for judging whether or not data collected in a data collection step shows a specific change by a management method using a management value. If the data analysis process determines that the data shows a specific change, the data analysis process includes a collection frequency adjustment process that changes the data collection frequency. When an abnormal change indicating a sign of a decrease is shown, the collection frequency adjustment step can automatically increase the collection frequency of the management target data. For this reason, it is possible to reduce the rate of occurrence of troubles in which the user does not notice that the management target data greatly deviates from the standard until the next measurement. Also,
It is possible to shorten the period in which the standard is not met. In addition, when it is likely to deviate from the standard, the frequency of measurement collection will increase, so that data fluctuation can be grasped in detail,
It is possible to determine the very last time when it deviates from the standard.

According to the data collection method of the second aspect of the present invention, it is evaluated whether or not the collection frequency is an appropriate value based on whether or not the data analyzed in the data analysis step is within the range of the standard. If it is inappropriate, it includes a collection frequency evaluation step in which the collection frequency of the data collection step is corrected by the collection frequency adjustment step.Therefore, evaluate whether the current collection frequency is appropriate. , So as to increase the collection frequency. By providing the collection frequency evaluation step, if the collection frequency is inappropriate, the collection frequency can be automatically corrected. Thus, it is possible to solve a problem that data is continuously managed at an inappropriate collection frequency.

According to a third aspect of the present invention, the collection frequency evaluation step includes a step of evaluating whether the collection frequency is an appropriate value at least from the time when the data shows a specific change to the next data collection. Counts the number of times the
If this count satisfies a predetermined condition, it is determined that the collection frequency is not an appropriate value. Therefore, if the measured value of the management target data exceeds the range of the management value, the current collection frequency is inappropriate. Increase the value of the number of times determined to be one, otherwise increase the value of the number of times the current collection frequency is determined to be valid by one, and judge that the total number of times and the current collection frequency are inappropriate Depending on whether the rate
The validity of the collection frequency can be evaluated, and any inappropriate collection frequency can be corrected.

According to the data collection method of the fourth aspect of the present invention, instead of the collection frequency adjusting step of the data collection method of the first aspect, when it is determined in the data analysis step that the data shows a specific change, Includes a collection frequency change notification step that notifies the user to change the data collection frequency in the data collection process, so even if a person performs measurement and collection of managed data, notifies the relevant person to change the collection frequency can do. As a result, the frequency of data collection in the data collection process is increased.

[Brief description of the drawings]

FIG. 1 is a functional block diagram according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram showing an example of a management method for detecting an abnormal change in management target data.

FIG. 3 is an explanatory diagram of a table structure of a database used in a data collection step.

FIG. 4 is an explanatory diagram of a table structure of a database used in a data analysis step.

FIG. 5 is an explanatory diagram of a table structure of a database used in a collection frequency adjusting step.

FIG. 6 is an explanatory diagram illustrating an example of a change in a collection frequency in the first embodiment.

FIG. 7 is a functional block diagram according to a second embodiment of the present invention.

FIG. 8 is an explanatory diagram of a table structure of a database used in a collection frequency evaluation step.

FIG. 9 is an algorithm used in a collection frequency evaluation unit in the second embodiment.

FIG. 10 is a functional block diagram according to a third embodiment of the present invention.

FIG. 11 is an explanatory diagram showing a table structure of a database used in a collection frequency change notification step.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Data collection means 2 Data analysis means 3 Collection frequency adjustment means 4 Collection frequency evaluation means 5 Collection frequency change notification means 10 Computer 20 Measurement device 30 Database 40 Database 50 Database 60 Database 70 Database 70 Database f31, f32 Table fields f41 and f32 Ｆf44 Fields of a table of the database 40 f51 to f54 Fields of a table of the database 40 f61 to f65 Fields of a table of the database 60 f71 to f73 Fields of a table of the database 70 S1 to S4 Algorithm used in the collection frequency evaluation means 4 Each step

Claims (4)

[Claims] 1. A data collection step of measuring and collecting data to be managed, and the data collected in the data collection step are:
A data analysis step of determining whether or not a specific change has been shown by a management method using a management value, and when it is determined that the data has shown a specific change in the data analysis step, the data collection frequency of the data collection step is reduced. A data collection method including a collection frequency adjustment step of changing. 2. A data collection step of measuring and collecting data to be managed, and the data collected in the data collection step are:
A data analysis step of determining whether or not a specific change has been shown by a management method using a management value, and a collection frequency adjustment step of changing the data collection frequency of the data collection step, and the data analysis step Evaluate whether the analyzed data is within the range of the standard to determine whether the collection frequency is an appropriate value or not.If inappropriate, correct the collection frequency of the data collection step by the collection frequency adjustment step. A data collection method that includes a collection frequency evaluation step. 3. The collection frequency evaluation step includes, in order to evaluate whether or not the collection frequency is an appropriate value, at least from the time when the data shows a specific change to the time when the next data is collected, the data is compared with the standard. 3. The data collection method according to claim 2, wherein the number of deviations is counted, and if the count satisfies a predetermined condition, it is determined that the collection frequency is not an appropriate value. 4. A data collection step of measuring and collecting data to be managed, and the data collected in the data collection step are:
A data analysis step of determining whether or not a specific change has been shown by a management method using a management value, and when it is determined that the data has shown a specific change in the data analysis step, the data collection frequency of the data collection step is reduced. A collection frequency change notification step of notifying the user to change the data.