Chaughule, Thorat - Statistical Analysis/Design of Experiments
the response function. The second-order model is the most applied empirical model.
This kind of model showing a suitable quality is required to obtain the response function.
The most commonly applied designs are three-level factorial designs, central composite
designs, Box-Behnken designs, and Doehlert matrixes. The choice of the response varia-
ble is an important step in the optimization.
Together these experimental designs can be used to determine process conditions
for achieving the target value and to identify variables that can be controlled to reduce
variation in key performance characteristics. It is important to note that these new phi-
losophies and experimental strategies have been highly promoted primarily because of
the renewed emphasis on product quality improvement as a means to attain competitive
advantage.
All the procedures recommended, including screening experiments, factorial expe-
riments, and specialized designs, can be used in suitable quality-improvement settings.
The responses of interest in quality-improvement studies include both measures of
product or process location and measures of dispersion. Many times the goal may not
be to find optimal points (maxima or minima) in the response surfaces that coincide, but
rather to locate flat regions that give stability of the responses. This is particularly true
in product design (robustness) and process design (process control).There are various
software commercially available for design of experiments such as start ease, stata-
graphics etc.
The reason for the recent popularity of these statistical and experimental strategies
is the competitive environment of today’s marketplace in many manufacturing and food
process industries. The statistical design techniques discussed here in this chapter can
be used to determine desired factor settings so that a process average or a quality cha-
racteristic of key product properties are close to the target (on aim) and the variability is
as small as possible.
REFERENCES
Akhnazarova, S., Kafarov, V., 1982, Experimental optimization. In: Chemistry and Chemical
Engineering. Moscow: Mir Publishers.
Andrade, I., Flores, H., 2004 , Optimization of spray drying of Roselle extract, In: Drying
2004 – Proceedings of the 14th International Drying Symposium, Vol A. State University of
Campinas, Sao Paulo, Brazil, pp. 597 - 604.
Antony, J., 2003, Design of Experiments for Engineers and Scientists, Elsevier Science &
Technology Books.
Araujo, P.W., Brereton, R.G., 1996, Experimental design. 2. Optimization, Trends Anal Chem
15(2), pp. 63 – 70.
Birchal, V.S., Passos, M.L., Wildhagen, G.R.S., Mujumdar, A.S., 2005, Effect of spray-dryer op-
erating variables on the whole milk powder quality, Drying Technology, 23(3), pp. 611-636.