Evolutionary Operations, first described by George E. P. Box and Norman R. Draper in in their book Evolutionary Operations – A statistical method for process improvement, New York: John Wiley and Sons, 1969.
EVOP is Continuous Improvement + Design of Experiments.
Basic idea is to replace the routine operation of a process by continuous and systematic plan of slight adjustments of the control variables. The effects of the adjustments are then evaluated just as with DOE. The process is then shifted in the desired direction of improvement. In many product and service processes it is impossible or very expensive to do DOE, especially where trials can be disruptive or the process owner would let you have the necessary time, materials, labor to run your experiments. So rather than running experimental production runs you use actual production by shifting off the base point left, right, up, down all within "spec".
The procedure goes like this:
- Choose a base point, a place to start, typically the center of the control variables specification ranges.
- Evaluate the objective function at the base point and determine a set of regularly spaced points around the base point. For 2 variables, form a square with the base point at the center. For 3 variables, form a cube, etc.. That is, conduct a DOE or Taguchi using the live process.
- Then, shift the process center to the point that has the largest value of the objective function.
- Use this point as the new base point.
- Repeat steps (2) and (3) until no further improvement occurs.
EVOP can be slow, months instead of weeks, so you need some patience and structure. For organizations that embrace the method it becomes a basic part of the continuous improvement scientific management culture.