Automation, Reliability, in a Factory Setting
Abstract
James and Fred discussing how factory equipment automation can create large amounts of data that may or may not be useful.
Key Points
Join James and Fred as they discuss the ability to collect data does not always translate into useful information to improve your equipment reliability or maintenance program.
Topics include:
- Before gathering more data make sure your maintenance fundamentals are in place and working well
- Practice deliberate data collection rather than collect and hope you can use it
- The essential need to fully understand failure modes/mechanisms and what your sensor data means
Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques, to field data analysis approaches.
/
RSS Feed
Show Notes
Completing The Right Maintenance At The Right Time With CBM & PdMÂ James’ Maintenance & Reliability article
58 Understanding Thermal Growth with John Lambert (a Rooted in Reliability episode)
The IoT is a little scary and for the consumer, where is the value? I recently noted that my iPhone maps knows where I am going, without putting in a destination. I live near Flint, but go to Church on Sunday in East Lansing. When I leave home, a message pops up giving the distance and information. When I leave for home, a few hours later, another similar message pops up. How does it know? What value is this for a routine trip where I know the weather and distance?
Another comment. I worked in the Automobile industry and saw how lots of data was collected and stored without analysis, or very limited analysis. Why? First, the shear volume of information is intimidating. Second, many engineers and analysts are working on laptops with very limited memory and processing speed. Because some data drives millions of dollars in cost, like warranty, there are whole departments combing the data for issues. Other data sources, like telematic vehicle information, receives very little analysis. Why? Primarily because most people don’t know what questions they need answered or how together the answers.
I spend several years creating ways to analyze and present the results for several types of telematics data. Basically, I took statistical numeric and graphical methods used with small data and applied it to the Big Data.