Chaughule, Thorat - Statistical Analysis/Design of Experiments
8.1. INTRODUCTION
Scientific experiments are important in research laboratories of universities and al-
so in the engineering laboratories of industries. Experiments in processing companies
are often conducted in a series of trials or tests which produce quantifiable outcomes.
For continuous improvement in product/process quality, it is fundamental to under-
stand the process behavior, the amount of variability and its impact on processes
(Mason et al., 2003).
Quality and productivity with process economics are characteristic goals of industri-
al processes, which are expected to result in superior goods that are highly sought by
consumers and that yields profit for firms that supply them. Recognition is now being
given to the necessary link between scientific study of industrial drying process and the
quality of dried products produced. The stimulus for this recognition is the intense in-
ternational competition among companies selling similar dehydrated products to a li-
mited consumer group. Competition demands that a better product need to be produced
within the limits of economic realities. Better food products are initiated in academic
and industrial research laboratories, made feasible in pilot studies and new-product re-
search studies, and checked for adherence to design specifications throughout
processing stages (Mason et al., 2003). All of these activities require experimentation
and the collection of data. Experimental design is a critically important tool in the food
engineering world for improving the performance of a dehydration or formulation
process. It also has extensive applications in development of novel drying techniques.
Design of Experiments (DOE) is a statistical technique introduced in the 1920s by Sir
Ronald Fisher in London. His initial experiments were concerned with determining the
effect of various fertilizers on different plots of land. The final condition of the crop was
not only dependent on the fertilizer but also on a number of other factors (such as un-
derlying soil condition, moisture content of the soil, etc.) of each of the respective plots.
Fisher used DOE which could differentiate the effect of fertilizer and the effect of other
factors. Since then DOE has been widely accepted and applied in many research discip-
lines (Antony, 2003).
The applications of experimental design techniques in process development can re-
sult in
- Improved process yields and product quality.
- Reduced variability and closer performance to target requirements.
- Reduced number of steps and development time.
- Improved process economics.
- Increased understanding of the relationship between key process inputs and
output (s). (Montgomery, 2003)
An experiment is a series of tests, called runs, in which changes are made in the in-
put variables in order to identify the reasons for changes in the output response (Box et
al, 2005). The term scientific study indicates a process of objective investigation which
makes sure that valid conclusions can be drawn from experimental study. Figure 8.1
signifies that statistics is imperative in every step of data collection and analysis, from
initial problem recognition and definition of objectives to the drawing of final conclu-
sions. Figure 8. 1 distinguishes two types of studies: experimental and observational. In
experimental studies the variables of interest often can be controlled and fixed at prede-
termined values for each test run in the experiment. In observational studies many of