sory quality can be done by the use of sensory descriptive analyses based
on the perceptions of trained panelists. It is also assumed for both the
panelists and the consumers that their perception is a function of three
factors, which were selected in order to maximize the variation of the
pea quality. The three factors, which form an experimental design, were
size (4 levels), content of sucrose (2 levels) and surface color (2 levels,
L-value from the HunterLab tristimulus Lab-measurements). After
screening 42 batches of peas from the Danish 1993 harvest, 16 pea sam-
ples were selected according to this design. However, two of the 16 com-
binations were not available, so it was not possible to reach a completely
balanced design. But still it is possible to estimate all the main effects
and the two factor interactions considered most important.
Two consumer surveys were conducted in Denmark, a qualitative
study in the form of four focus group interviews followed by a quanti-
tative study including an in-home product test. The purpose of the focus
groups was to gain information on consumer attitudes and behavior in
relation to vegetables in general and specific knowledge of consumer
needs and wants regarding deep frozen green peas, in QFD terminology
listen to the voice of the customer(Griffin and Hauser, 1993). The re-
sults were used for development of the questionnaire for the quantita-
tive study. Consumer needs and wants regarding good sensory quality
of peas are listed in the left part of Figure 11.5.
The results from the consumer evaluations as well as the evaluations
by the trained panelists were analyzed by using General Linear Models
(GLM for Unbalanced ANOVA, SAS) with the evaluations as depen-
dent variables and the experimental factors as the independent variables;
the interactions considered most important were also modeled. For all
significant effects the differences were analyzed in detail in order to ex-
plain the direction of the significant effects. The results from the con-
sumer study and the sensory descriptive study are illustrated in Figure
11.5, the left part of the central relationship matrix and the upper part
of the roof of the House of Quality, respectively.
By use of matrix algebra, the consumer requirements regarding good
sensory quality are translated into sensory attributes measured by trained
panelists. The first element of the right part of the central relationship
matrix is a result of the sum of the multiplication of each element in the
corresponding row with each element in the corresponding column
(roof). In this case, the consumers’ perception of the ideal smooth sur-
face of peas is related to the size of the peas, where size 1 is preferred,
and to the content of sucrose, where the high level (H) is preferred. The
Example—House of Quality for Strategic Peas 217