公开(公告)号：WO2021099678A1

公开(公告)日：2021-05-27

申请号：PCT/FI2020/050739

申请日：2020-11-09

Applicant: ELISA OYJ

Inventor： HEIKKILÄ, Rasmus

IPC: G05B23/02

Abstract: A computer implemented method of analyzing measurement results of a target system, such as an industrial process or communication network. The method comprises receiving (301) a data sample comprising a plurality of variables representing the measurement results, detecting (302) that the data sample is an anomalous sample using an anomaly detection model, processing (303) the data sample by applying an imputation model to selected subsets of variables of the data sample to obtain imputed samples, and applying the anomaly detection model to the imputed samples, determining (304) anomalous variables of the data sample based on results from the processing of the data sample, and outputting (305) the anomalous variables of the data sample for management operations in the target system.

公开(公告)号：WO2022263713A1

公开(公告)日：2022-12-22

申请号：PCT/FI2022/050389

申请日：2022-06-06

Applicant: ELISA OYJ

Inventor： YLÄ-JARKKO, Kalle ,  HEIKKILÄ, Rasmus ,  LISKI, Antti ,  HILTUNEN, Eero ,  SALMENKAITA, Jukka-Pekka

IPC: H01L21/66 ,  G05B19/00 ,  G06F17/00

Abstract: Monitoring of a semiconductor manufacturing process. Wafer measurement data is obtained (301); Zernike polynomials are fitted (302) to the wafer measurement data to obtain representation of respective wafermap patterns; a knowledgebase of wafermap patterns is built (303) based on the respective coefficients of the Zernike polynomials; the wafermap patterns of the knowledgebase are grouped (304) to wafermap pattern groups based on the respective coefficients of the Zernike polynomials; and at least some of the wafermap pattern groups of the knowledgebase and the respective coefficients of the Zernike polynomials are used (305) for analyzing new wafer measurement data for the purpose of monitoring the semiconductor manufacturing process.

公开(公告)号：WO2022129677A1

公开(公告)日：2022-06-23

申请号：PCT/FI2021/050820

申请日：2021-11-29

Applicant: ELISA OYJ

Inventor： UURTIO, Viivi ,  YU, Qi ,  HEIKKILÄ, Rasmus ,  LUNDÉN, Petteri ,  LISKI, Antti

IPC: G06F17/16 ,  G06F17/18 ,  G06N3/00

Abstract: Analyzing measurement results of a target system. The analysis is performed by receiving (311) a first matrix (301) comprising first measurement results of the target system; training (312) a matrix decomposition model with the first matrix (301) to obtain a third matrix (303) of normal or stable measurement results and a fourth matrix (304) of anomalous or unstable measurement results; receiving (313) a second matrix (306) comprising second measurement results of the target system, wherein the second measurement results are later measurement results compared to the first measurement results; selecting (314) from the third matrix (303) a subset (307) that matches with the second matrix (306); subtracting (315) the selected subset (307) from the second matrix (306) to obtain a fifth matrix (308); outputting (316) the fifth matrix (308) or information derived from the fifth matrix (308) for the purpose of evaluating performance of the target system.

公开(公告)号：WO2022117912A1

公开(公告)日：2022-06-09

申请号：PCT/FI2021/050784

申请日：2021-11-18

Applicant: ELISA OYJ

Inventor： SALMENKAITA, Jukka-Pekka ,  HEIKKILÄ, Rasmus

IPC: G05B19/418 ,  H01L21/67

Abstract: Monitoring and control of a semiconductor manufacturing process. Candidate rules for the statistical process control of the semiconductor manufacturing process are automatically generated based on production data and quality data; the candidate rules are submitted for evaluation; and one or more of the candidate rules are conditionally taken into use in the statistical process control of the semiconductor manufacturing process.

公开(公告)号：WO2022090609A1

公开(公告)日：2022-05-05

申请号：PCT/FI2021/050679

申请日：2021-10-13

Applicant: ELISA OYJ

Inventor： HEIKKILÄ, Rasmus ,  LISKI, Antti

IPC: G06N20/20

Abstract: A computer implemented method of building an ensemble of anomaly detection models for analyzing measurement results of a target system. The method is performed by obtaining (301) a plurality of measurement result samples comprising a plurality of variables of the target system; processing (302) the plurality of measurement result samples with a plurality of anomaly detection models to obtain a plurality of anomaly scores for each measurement result sample; determining (303) disagreement scores for the measurement result samples based on the respective plurality of anomaly scores; and selecting (304) measurement result samples with a disagreement score that fulfils predefined criteria for evaluation by an expert to obtain confirmed labels for building (305) the ensemble of anomaly detection models.

公开(公告)号：WO2021250313A1

公开(公告)日：2021-12-16

申请号：PCT/FI2021/050402

申请日：2021-06-02

Applicant: ELISA OYJ

Inventor： SALMENKAITA, Jukka-Pekka ,  HEIKKILÄ, Rasmus

IPC: H04L12/24 ,  H04L41/0636 ,  H04L41/0686 ,  H04L41/069

Abstract: A computer implemented method of monitoring and controlling a target system, such as a communication network or an industrial process. The method includes receiving (301) information about anomalies in operation of the target system detected by an automated anomaly detection mechanism; automatically determining (302) certainty characteristics of the detected anomalies; submitting (303) detected anomalies to expert evaluation in priority order determined based on the certainty characteristics; and adjusting (304, 305) the determination of certainty characteristics of the detected anomalies and/or the automated anomaly detection mechanism based on results of the expert evaluation.

公开(公告)号：WO2021219932A1

公开(公告)日：2021-11-04

申请号：PCT/FI2021/050307

申请日：2021-04-23

Applicant: ELISA OYJ

Inventor： HEIKKILÄ, Rasmus ,  LISKI, Antti

IPC: G06N3/04 ,  G06N3/08

Abstract: A computer implemented method for obtaining an autoencoder model for the purpose of processing metrics of a target system. The method comprises obtaining (301) a data set comprising metrics associated with the target system, the data set being intended for training the autoencoder for processing further metrics of the target system; masking (302) the data set with a predefined mask configured to exclude certain parts of the data set; using (303) the unmasked parts of the data set for training the autoencoder; masking (304) reconstructed data from the autoencoder with the same predefined mask; using (305) reconstruction error of the unmasked parts of the reconstructed data to update parameters of the autoencoder to obtain autoencoder model; using (306) the masked parts of the data set for testing the autoencoder model; and providing (307) the autoencoder model for processing further metrics of the target system.