
87
6
Diagnosing Variation in
Materials Processing
“I’ve looked at the past 12 months of potline data, Rob,” said Tony, the
statistician whom Rob had called for help. “There’s a lot there, and I can’t
interpret why many things have happened. That’s for you. However, I can
say that there have been 10 instances where the process has gone into an
out-of-control state for a period of a week or more: either temperature, AlF3%,
pot voltage, noise (mV), AE (anode effect) frequency (AEF), anode problems,
unscheduled anode changes (USCs), Fe%, Si%, or some combination of these.”
“Okay, but things happen in production, Tony,” Rob said. “We have to roll ...