One factor at a time testing varies one input, picks its best value, then moves to the next. It feels intuitive, but it needs more runs for the same precision, explores only a cross-shaped slice of the factor space and misses interactions entirely, so the optimum it finds can sit far from the real one. Among quality engineers, calling an experiment OFAT is a mild insult.
That's OFAT dressed up in a spreadsheet. You changed pressure, then power, and never ran them together.
Also heard as
- one factor at a time
- one-variable-at-a-time
- OVAT
Related terms
DOE
Design of experiments: planning a set of runs that vary several factors at once so their effects can be separated.
Interaction
Situation where the effect of one factor depends on the level of another, so their combined effect is not simply additive.
Factorial design
Experiment that runs every combination of the chosen factor levels, so each main effect and interaction can be estimated.
Knob
Any adjustable process parameter on a tool, such as RF power, pressure, gas flow, temperature or time.