This study sets out to establish a simple online quality control measurement for the manufacture of a cereal product. The purpose was to enrol consumers to dictate the quality control parameters. A methodology was employed whereby a variety of products were developed using an experimental plan based on consumer opinions and manufacturing capabilities. A trained panel profiled each of these products using descriptive analysis. In addition, physical and chemical analyses were carried out on the products. Consumers then evaluated these cereal products for acceptability at breakfast time. The analytical results, trained panel data and consumer data were then combined using regression and partial least squares regression analysis to form a preference map. The results indicated that a single physical measure could predict 73% of the time whether the product was to be acceptable or not to the consumer. If the ratings for the batch were outside this acceptability reading, then the products were to be placed on hold for further analysis. This simple measurement can now be used online to determine if a specific batch of cereal meets consumers' expectations of product quality.
[1]
T. Dijkstra.
Some comments on maximum likelihood and partial least squares methods
,
1983
.
[2]
Herbert Stone,et al.
Quantitative Descriptive Analysis (QDA)
,
1992
.
[3]
N. Gains,et al.
The repertory grid approach
,
1994
.
[4]
Robert C. Hootman.
Manual on descriptive analysis testing for sensory evaluation
,
1992
.
[5]
H. J. H. MacFie,et al.
Measurement of Food Preferences
,
1994
.
[6]
Mina R. McDaniel,et al.
USE OF CONTRASTS FOR THE EVALUATION OF PANEL INCONSISTENCY
,
1990
.
[7]
Mina R. McDaniel,et al.
THE PANELIST EFFECT – FIXED OR RANDOM?
,
1988
.