Machine Learning as the backbone for Predictive Failure Analytics of Devices-Ramki Pitchuiyer -02/28/2017 - 8:30am

Event Information
Event Topic: 
Machine Learning as the backbone for Predictive Failure Analytics of Devices
Event Date: 
02/28/2017 - 8:30am
Event Location: 
Sunnyvale City Council Chambers- 456 West Olive Ave., Sunnyvale
Speaker Information
Event Speaker: 
Ramki Pitchuiyer
Event Speaker Title: 
Founder and CEO
Event Speaker Company: 
Event Speaker Bio: 

Ramki Pitchuiyer is currently the founder and CEO of eKryp, a stealth-mode start up focused on using machine learning and deep analytics to increase the uptime of critical machines in the industrial sub segments such as medical devices, biotech equipment, and money machines. Prior to eKryp, Ramki co-founded Paloras in 2011, which focuses on helping clients transform business processes in Product Life Cycle Management, Supply Chain and Customer Experience. In 2013 and 2014, Paloras was awarded Top 100 Businesses in the Bay Area by the San Francisco Business journal.

Ramki began his career as a programmer at Tata Unisys and Unisys, and later moved to management consulting at Ernst & Young. He then held senior management roles at Oracle and Ernst & Young prior to venturing out with Paloras and eKryp. At Oracle, he led consulting solution organization spanning several major industries: High Tech, Aerospace & Defense, Automotive, Industrial Manufacturing, CPG, and Oil & Gas. Prior to that he was engaged in major system integration and implementation for Fortune 500 companies and managed multiple $60M+ global programs with teams across continents.

Ramki holds an MBA from the Anderson Graduate School of Management at UCLA, and a B.S in Electrical and Electronics Engineering from BITS, Pilani in India. Ramki has been mentoring youth in science, technology and mathematics since 2007 as FIRST robotics coach, math coach and Western Regional robotics volunteer.

Event Details
Event Details: 

Unprecedented growth in IoT and connected devices is producing a variety of challenges and opportunities. The advent of machine learning along with Big Data technology, coupled with efficient cloud based deep data storage mechanisms and parallel processing frameworks, has allowed us to come up with ways to increase uptime of critical machines.

The presentation details the journey into using machine learning for predictive failure analytics and lessons learnt to drive such a solution within medical and biotech enterprises. With an increasing number of products that have embedded sensors and the ability to communicate, it is becoming imperative to monitor the usage, condition, and operations of these assets in real time. Furthermore, the rapid growth in data for offline and real time analytics enables failure prediction, preventive maintenance, and optimization of operations.

We will discuss how machine learning can enhance the customer experience and lower service costs. The discussion will take a case study and discuss how enterprises can avoid unscheduled downtime and increase uptime through data driven diagnostics. As an added benefit, such a solution would also optimize parts inventory plan, improve overall service operations, and increase the Overall Equipment Effectiveness (OEE).