PREDICTIVE ANOMALY DETECTION
    1.
    发明申请

    公开(公告)号:US20180157888A1

    公开(公告)日:2018-06-07

    申请号:US15370709

    申请日:2016-12-06

    Abstract: Systems, methods, and computer-readable storage media are provided for an embedded, scalable, predictive tool capable of detecting in-field anomalies and trends in advance of productivity losses on single devices, device clusters, and/or multi-cluster architectures. In-field and in real-time, sets of barcode signal sequences associated with respective barcode symbols are collected in time series (that is, at successive time intervals). A quality index measure in computed for each of the barcode signal sequence sets such that each quality index measure is associated with a barcode symbol. Patterns among the sets are identified therefrom and compared to barcode symbol patterns that are known to be associated with particular trends or anomalies and appropriately classified as such.

Patent Agency Ranking