There are many processes where some type of feedback-control scheme would be preferable to a control chart. For example, consider the process of driving a car, with the objective of keeping it in the center of the right-hand lane (or equivalently, minimizing variation around the center of the right-hand lane). The driver can easily see the road ahead, and process adjustments (corrections to the steering wheel position) can be made at any time at negligible cost. Consequently, if the driver knew the relationship between the output variable (car position) and the manipulated variable (steering wheel adjustment), he would likely prefer to use a feedback-control scheme to control car position, rather than a statistical control chart. (Driving a car with a Shewhart control chart may be an interesting idea, but the author doesn’t want to be in the car during the experiment.)