Patent · US Active

System and method for predicting and avoiding hardware failures using classification supervised machine learning

US12066908B2 · kind B2 · utility

0Cited by
28References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJul 29, 2022
Grant dateAug 20, 2024
Priority date
Expiry dateOct 6, 2042

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N20/00
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

A hardware failure prediction and avoidance system executing on a unified endpoint management platform information handling system comprising a network interface device (NID) to receive failed operational telemetries for client devices including power and software analytics, and event viewer error logs, and an error associated with a hardware type, a processor to execute a classification supervised learning algorithm on the failed operational telemetries to determine that a hardware or software configuration will likely co-occur with a future error of the hardware type, identify the hardware or software configuration as problematic, the processor to identify the problematic system configuration in a current operational telemetry by correlating the current operational telemetry with a portion of the failed operational telemetries, and the NID to transmit a recommendation to the first client device to adjust the problematic system configuration in order to avoid occurrence of the error at the first client device.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.