In this lesson, we're going to take a look at an experiment that studied the effect of several factors on wine quality using a 2^8-4 fractional design. The case is the following: Harry owns a vineyard and winery. He grows several varieties of grapes and produces wine. He is always experimenting with the wine-making variables to improve his products. This problem describes the experiment conducted for the 1985 Pinot Noir.

Eight variables were studied in a 2_IV^8-4 fractional design with 16 runs. The levels of the factors are described in the table:
Factor A, the pinot noir clone – Pommard or Wandenswil.
Factor B is the oak type - allier or troncais. Factor C is the age of the barrel - old or new.
Factor D is the yeast/skin contact - Champagne or Montrachet.
Factor E, the fermentation temperature - low or high.
Factor F, the presence of the whole cluster - none or 10 %.
Factor G, the presence of stems – none or all. And finally, factor H, the barrel toast - light or medium.

The wine was tasted by a panel of experts. Each expert ranked the 16 samples of wine tasted, with rank one being the best.