Tuesday, July 13, 2010

Who drives machine vision in your plant?

It’s my experience that responsibility for machine vision typically falls somewhere between Manufacturing Engineering and Maintenance. What decides which side of the divide it ends up seems to depend on whether the systems are smart cameras or PC-based. Smart cameras get classed as sensors and so belong to Maintenance while anything with a PC is Engineering.
This isn’t wrong but today I want to remind everyone that there’s a third party involved: the Quality department.

Quality is often the driving force behind greater use of machine vision – we implement systems to catch any instances of nonconforming product and protect our customers from the same – but we miss the opportunity to incorporate a Quality perspective into our vision projects.

Quality isn’t just about inspection. In fact 21st Century Quality thinking is all about avoiding waste by controlling and reducing variation in the process. There’s a huge emphasis on collecting and analyzing data as a way to get to root causes. It’s the Edwards Deming Plan-Do-Check-Act cycle, and machine vision should be providing the “Check” function.

But machine vision, as you well know, can do much more than just check: it’s a powerful, but underutilized, tool for gathering process data. It should be employed for much more than simple measurements; it should be the input into a statistical approach to quality improvement.

This is all getting outside the realm of manufacturing engineering, which brings me to the point of today’s sermon: talk to your Quality people about what machine vision can do for them. Educate them in the possibilities for data capture and trend monitoring and they’ll come back to you with a wish list as long as your arm.

Machine vision: a Quality tool that’s owned by Engineering.

No comments: