Return to Blog

Blog

FAQ

FAQ: FACTORY SENTINEL FOR INDUSTRIAL ROBOTS

MAY 8, 2018 - MATTHEW BUZZA, DATA SCIENTIST

Q: What is Factory Sentinel for Industrial Robots?

A: Factory Sentinel for Industrial Robots is a predictive maintenance solution for industrial robots. Its purpose is to detect early signs of failures and ultimately allow companies to predict when a robot is going to fail, so they can perform maintenance ahead of time and prevent costly downtime.

Q: What does Factory Sentinel for Industrial Robots do?

A: Factory Sentinel for Industrial Robots takes motor, current, position and speed data and runs it through a machine learning algorithm to model the baseline behavior of a robot. It then assesses how different current behavior is from the baseline, so personnel can visualize robot degradation.

Q: How does Factory Sentinel for Industrial Robots work?

A: Once the solution is set up, it collects about 30 data files from the robot and uses it to model the robot's normal behavior as well as the relationships between data signals. Over time, the solution details how different the asset is from when data was first collected.

In addition to displaying the robot’s overall health indicator, the solution provides joint information to make problem areas easily identifiable. Factory Sentinel for Industrial Robots also enables personnel to visualize statistical features of the motor current and take a deeper look at raw signals to understand robot features and compare motor current signals.

Q: What is the difference between Factory Sentinel for Industrial Robots’ alerts and reports?

A: Real-time alerts can be configured for multiple event types, but the solution generally sends out a text or email whenever a robot is about to fail or crosses the threshold of abnormality.

Reports can be sent out weekly, biweekly or monthly. They are customizable and typically show the factory’s five to 10 worst robots, as well as the robots that have changed the most within that time frame. The reports also show overall health trends and give personnel an in-depth look at each robot.

Q: What kind of robots does Factory Sentinel for Industrial Robots analyze?

A: The solution targets medium to large articulated arm robots, often referred to as six-axis industrial robots.

Q: Does Factory Sentinel for Industrial Robots work for all types of robots?

A: As long as the necessary data is available, the solution can work for all articulated arm robots.

Q: What kind of failures can Factory Sentinel for Industrial Robots predict?

A: The solution can predict mechanical failures that happen over time. Instantaneous failures cannot be predicted.

Q: Does Factory Sentinel for Industrial Robots collect data in real time?

A: Mechanical failures do not happen instantaneously. For that reason, Factory Sentinel for Industrial Robots does not need to collect 24/7 to predict failures. Instead, it collects data every 12 or 24 hours.

Q: Which industries benefit from Factory Sentinel for Industrial Robots the most?

A: Automotive manufacturers are the main beneficiaries of Factory Sentinel for Industrial Robots. However, other industries that experience downtime due to robot failures may also benefit from the solution.

Q: Who can use Factory Sentinel for Industrial Robots?

A: Any maintenance technician or manager can interpret the health information the solution provides. Factory Sentinel for Industrial Robots doesn’t require any prior data science experience, as the data analysis algorithm is already built into the solution.

Q: What kind of infrastructure is needed to run Factory Sentinel for Industrial Robots?

A: A designated IT specialist is needed to set up a server for the solution. The company will also need to retrieve robot data from controllers and set up the data collection so data can be pushed to the server.

Matthew Buzza

About the Author

Matthew Buzza is a Data Scientist for Predictronics Corp. with 5+ years of work experience in software design and development, as well as predictive maintenance solutions for machine tools, industrial robots, combustion engines and other vehicle systems. Matt co-developed Predictronics’ Factory Sentinel platform, a data analysis and visualization application used to monitor and prevent health issues within industrial robots. Matt has a Master of Science in Mechanical Engineering from the University of Cincinnati, where he studied intelligence maintenance systems at the NSF I/UCRC IMS Center founded by Professor Jay Lee. Connect with him on LinkedIn.

For more information and weekly updates, follow Predictronics:

Monitor your industrial robots today!

REQUEST A DEMO