Info

6B = Cost of goods less than 140.

CC = Cost of goods less than 130.

dD = Cost of goods less than 120.

eE = Perceived spiciness greater than 65.

6B = Cost of goods less than 140.

CC = Cost of goods less than 130.

dD = Cost of goods less than 120.

eE = Perceived spiciness greater than 65.

ing for a product that must be perceived as highly spiced (column E). Optimization yields the likely formulations, which are then used to estimate the likely response profile on all of the attributes. If the model includes attributes derived from physical measures, then one can calculate the profile of physical measures corresponding to the optimal formulation. This link between physical measures and ratings can be used to develop a quality control system (see below).

Fitting a Predesignated "Goal" Profile

The models allow one to prescribe a desired goal profile of subjective attributes, and then discover the particular combination of physical variables which generates that desired goal profile. Rather than optimizing a single response (eg, overall liking), the optimizer system varies the independent variables until the expected response profile matches as closely as possible the predesignated goal profile. (The closeness of fit can be expressed either in absolute terms, or in percentage terms. Percentage for "closeness of fit" is preferable when the units of measurement for the response attributes differ from attribute to attribute). Table 20 shows the solution, for two goal profiles. The first profile comes from consumers who evaluated a competitor product that the researcher wishes to "match" using the set of ingredients. The second goal profile was generated at the plant by instrumental probes which recorded the characteristics of each batch as it was produced. The objective was to estimate consumer reactions to this batch, using the instruments as quality probes.

Profile fitting enables the investigator to estimate the likely profile for one data set, given the profile for another data set. Each target profile constitutes a goal. It is possible to:

1. Relate Difference Sources of Data. Given a target profile from one source (eg, instrumental measures) the investigator can estimate the corresponding attribute profile which would emerge from another source of data for the same product (eg, consumer sensory ratings).

2. Program a set of "probes" (in-line) in a processing plant to respond to the formulation as consumers would. Given the profile of responses to a batch obtained from the probe, the quality assurance engineer can estimate the likely formulation that would have yielded the profile, and then from that formulation estimate the likely consumer ratings.

Table 20. Example of Goal Fitting

Consumer goals

Instrumental goals

Table 20. Example of Goal Fitting

Consumer goals

Berry Boosters

Berry Boosters

Acai, Maqui And Many Other Popular Berries That Will Change Your Life And Health. Berries have been demonstrated to be some of the healthiest foods on the planet. Each month or so it seems fresh research is being brought out and new berries are being exposed and analyzed for their health giving attributes.

Get My Free Ebook


Post a comment