ML for circuit quality assurance

Overview:

In today’s high-volume production environment for cutting-edge electronic systems such as networking equipment, component complexity and usage quantity both increase at an unprecedented pace. At the same time, quality requirements for these systems must be continuously improved as even small glitches can have major consequences. To quickly detect and identify potential abnormal component behaviors during either system manufacturing or customer application, it is becoming more and more challenging to rely on manual data processing because of the high quantity of components used, which can be in the millions each single day and with billions of components already in deployed systems. Juniper Networks’ Component Engineering team is currently collaborating with iCAS Lab on a joint research project to develop a machine learning approach to enable autonomous data analytics, with the objective to monitor quality performance, enable early failure detection, and accelerate failure resolution.

​

Sponsors:

• Juniper Networks, Inc.

​

Student Contributors:​

• Brendan Reidy

• Brenden Chavis