As a follow-on to my May column about machine safety urgency (www.ControlDesign.com/hardway), let's pursue the training issue that closed the piece.
The tragic and fatal train-crossing accident that I wrote about spawned questions about the origins, amount and usefulness of the controls/safety training that machine automation designers, engineers, technicians and others receive from or through their employers or seek on their own. This would include procedures that must be followed and can't be subject to shortcuts, regardless of the situation at hand.
Overall, it seems we've made good progress toward making ever-more complex and powerful industrial machinery capable of high-quality and flexible production in a very safe manner.
"Overall," however, is a dangerous word. Just like "average," it usually includes some extremes of good and bad. Lately, with train crashes, deep-water drilling disasters, fast-moving Toyotas and more, the bad extremes show a troubling, high-impact frequency.
This made me wonder if performance evaluation, and any subsequent reward, is tied to adherence to these learned skills and techniques. What unfolds when the prescribed training doesn't reflect what happens, what often has to happen, on the production floor?
I had a flashback to my manufacturing days, and I'm hardly alone when I recall stories about electricians and technicians bypassing mandated procedures to restore some breakdown quickly. Sometimes it's self-initiated action. Other times it's pressure from an equally beleaguered floor supervisor. Some companies would view adherence to proper procedures in those circumstances unfavorably. The hard, unspoken reality here is the rewards are based on output, not procedural adherence.
For machine builders it's not as clear cut. With today's technology, your company can do everything right and build a highly automated machine that gives your customer a perfect harmony of productivity, flexibility and operator safety. Properly operated, that is.
Machine builders have highly varied levels of after-sale support, so some of you don't know much about what happens after commissioning and startup. During installation, you might not get a clear picture of the training the operators and technicians will have.
So, and I'm not trying to hang this on you, what role—if any—do you play in designing a system that can't be defeated in a manner that has potential for big trouble? Does it sometimes have to go beyond a defendable risk assessment and all the due diligence that entails?
I even hear a call to return to the use of proprietary control boards and code so the user can't disable critical sequences. I don't think that's the progressive answer we're after, but we should talk more about this. Your turn.