Data processing method, data processing device, and storage medium

A data processing method comprising:

• processing a substrate using a substrate processing device;
• acquiring a plurality of time series data acquired by the substrate processing device;
• acquiring evaluation values of the plurality of time series data;
• classifying each of the plurality of time series data into one of a plurality of classifications based on a corresponding one of the evaluation values;
• extracting based on a result obtained in the classifying time series data corresponding to one of the plurality of classifications and representing an abnormal evaluation value as extracted time series data;
• and
• classifying the extracted time series data into one of a plurality of clusters through clustering processing,
• wherein the substrate is a semiconductor wafer having a disc shape,
• wherein in the acquiring of the evaluation values, the evaluation values of the plurality of time series data are acquired through comparison between the plurality of time series data and reference data, and
• wherein, in the acquiring of the evaluation values, an evaluation value of the evaluation values in a first case is larger than an evaluation value of the evaluation values in a second case, a difference between values of the time series data and the reference data in the first case being larger than a difference between values of the time series data and the reference data in the second case.

A data processing method comprising:

• processing a substrate using a substrate processing device;
• acquiring a plurality of time series data acquired by the substrate processing device;
• acquiring evaluation values of the plurality of time series data;
• classifying each of the plurality of time series data into one of a plurality of classifications based on a corresponding one of the evaluation values;
• extracting based on a result obtained in the classifying time series data corresponding to one of the plurality of classifications and representing an abnormal evaluation value as extracted time series data;
• classifying the extracted time series data into one of a plurality of clusters through clustering processing;
• and
• displaying the extracted time series data and an evaluation value graph representing changes of the evaluation value of the extracted time series data in a switchable manner,
• wherein the substrate is a semiconductor wafer having a disc shape, and
• wherein the extracted time series data displayed in the displaying corresponds to an evaluation value designated by a cursor.

A data processing method comprising:

• processing a substrate using a substrate processing device;
• acquiring a plurality of time series data acquired by the substrate processing device;
• acquiring evaluation values of the plurality of time series data;
• classifying each of the plurality of time series data into one of a plurality of classifications based on a corresponding one of the evaluation values;
• extracting based on a result obtained in the classifying time series data corresponding to one of the plurality of classifications and representing an abnormal evaluation value as extracted time series data;
• classifying the extracted time series data into one of a plurality of clusters through clustering processing;
• and
• displaying the extracted time series data and an evaluation value graph representing the evaluation value of the extracted time series data classified into one of the plurality of clusters in a switchable manner,
• wherein the substrate is a semiconductor wafer having a disc shape, and
• wherein the extracted time series data displayed in the displaying corresponds to an evaluation value designated by a cursor.

A data processing method comprising:

• processing a substrate using a substrate processing device;
• acquiring a plurality of time series data acquired by the substrate processing device;
• acquiring evaluation values of the plurality of time series data;
• classifying each of the plurality of time series data into one of a plurality of classifications based on a corresponding one of the evaluation values;
• extracting based on a result obtained in the classifying time series data corresponding to one of the plurality of classifications and representing an abnormal evaluation value as extracted time series data;
• classifying the extracted time series data into one of a plurality of clusters through clustering processing;
• displaying the extracted time series data and an evaluation value graph representing changes of the evaluation value of the extracted time series data in a switchable manner;
• and
• generating a learning database in which cause and countermeasure information for the substrate processing device that corresponds to the time series data is assigned to a cluster corresponding to the extracted time series data,
• wherein the substrate is a semiconductor wafer having a disc shape.

A data processing method comprising:

• processing a substrate using a substrate processing device;
• acquiring a plurality of time series data acquired by the substrate processing device;
• acquiring evaluation values of the plurality of time series data;
• classifying each of the plurality of time series data into one of a plurality of classifications based on a corresponding one of the evaluation values;
• and
• matching target time series data that has been classified into a particular classification of the plurality of classifications in the classifying and that represents an abnormal evaluation value and time series data included in a learning database and classified into a plurality of clusters in the particular classification of the plurality of classifications, except time series data corresponding to any of the rest of the plurality of classifications other than the particular classification,
• wherein the substrate is a semiconductor wafer having a disc shape.

A data processing method comprising:

• processing a substrate using a substrate processing device;
• acquiring a plurality of time series data acquired by the substrate processing device;
• acquiring evaluation values of the plurality of time series data;
• classifying each of the plurality of time series data into one of a plurality of classifications based on a corresponding one of the evaluation values;
• and
• matching target time series data that has been classified into one of the plurality of classifications in the classifying and that represents an abnormal evaluation value and time series data included in a learning database and classified into a plurality of clusters in the one of the plurality of classifications,
• wherein the substrate is a semiconductor wafer having a disc shape,
• wherein cause and countermeasure information corresponding to the time series data is stored in the learning database, and
• wherein, in the matching of the target time series data, reading the cause and countermeasure information corresponding to the time series data is further included in a case in which a matching rate between the target time series data and at least one piece of the time series data included in the learning database is higher than a threshold.

A data processing method comprising:

• processing a substrate using a substrate processing device;
• acquiring a plurality of time series data acquired by the substrate processing device;
• acquiring evaluation values of the plurality of time series data;
• classifying each of the plurality of time series data into one of a plurality of classifications based on a corresponding one of the evaluation values;
• and
• matching target time series data that has been classified into one of the plurality of classifications in the classifying and that represents an abnormal evaluation value and time series data included in a learning database and classified into a plurality of clusters in the one of the plurality of classifications,
• wherein the substrate is a semiconductor wafer having a disc shape,
• wherein in the acquiring of the evaluation values, the evaluation values of the plurality of time series data are acquired through comparison between the plurality of time series data and reference data, and
• wherein, in the acquiring of the evaluation values, the evaluation value of the evaluation values in a first case is larger than the evaluation value of the evaluation values in a second case, a difference between values of the time series data and the reference data in the first case is larger than a difference between values of the time series data and the reference data in the second case.