Exploiting multiple mahalanobis distance metric to screen outliers from analogue product manufacturing test responses

Shaji Krishnan, Hans G. Kerkhoff

    Research output: Contribution to journalArticleAcademicpeer-review

    20 Citations (Scopus)
    643 Downloads (Pure)


    One of the commonly used multivariate metrics for classifying defective devices from non-defective ones is Mahalanobis distance. This metric faces two major application problems: the absence of a robust mean and covariance matrix of the test measurements. Since the sensitivity of the mean and the covariance matrix is high in the presence of outlying test measurements, the Mahalanobis distance becomes an unreliable metric for classification. Multiple Mahalanobis distances are calculated from selected sets of test-response measurements to circumvent this problem. The resulting multiple Mahalanobis distances are then suitably formulated to derive a metric that has less overlap among defective and non-defective devices and which is robust to measurement shifts. This paper proposes such a formulation to both qualitatively screen product outliers and quantitatively measure the reliability of the non-defective ones. The resulting formulation is called Principal Component Analysis Mahalanobis Distance Multivariate Reliability Classifier (PCA-MD-MRC) Model. The application of the model is exemplified by an industrial automobile product.
    Original languageUndefined
    Pages (from-to)18-24
    Number of pages5
    JournalIEEE design & test of computers
    Issue number3
    Publication statusPublished - 2013
    Event2011 International SoC Design Conference, ISOCC 2011 - Kyung Hee University, Seoul, Korea, Republic of
    Duration: 17 Nov 201118 Nov 2011


    • EWI-22724
    • CAES-TDT: Testable Design and Test
    • Reliability
    • Analogue
    • Test
    • METIS-293262
    • IR-83464
    • Outliers

    Cite this