fore a fixed value and is consequently not expressed explicitly in the
equations used. As long as the selected quality criteria are applied con-
sistently throughout the experiments, neither the type nor the value of
the criteria for acceptability is important for the development of the
model. Acceptability is subject to local and regional preferences; hence,
to ensure generic application, the model has to be independent of ac-
ceptability criteria. If the criteria of acceptability do not change during
the assessment period, the decrease in acceptability is completely de-
termined by the loss of those intrinsic product properties that contribute
to the perceived quality.
The plants were given a standardized pretreatment, stored in darkness
at six constant temperatures. Up to 21 days, quality was evaluated six
times. The experimental conditions of time and temperature were cho-
sen to simulate the conditions encountered during container transport to
medium and distant markets.
Time Effects
Loss of perceived quality over time is undoubtedly the result of a cas-
cade of biochemical reactions. Such cascades can often be approximated
by the sigmoid logistic curve. A logistic function is therefore likely to
be a good model. The formulation for the logistic curve has been slightly
adapted to enable the introduction of the effect of temperature in a way
suitable for the dynamic approach:
N (7.1)
where Nmaxis the number of plants in the batch, kis the rate of decrease
in acceptability, tis the time in days, Nis the number of acceptable
plants, N 0 is the initial number of acceptable plants. [(Nmax/N 0 )1] con-
stitutes a correction for the biological age of the plants at the start of the
experiment (Thai et al., 1990; Tijskens and Evelo, 1994c) and is inde-
pendent of the temperature applied in the subsequent treatment.
Temperature Effects
The rate kin Equation (7.1) represents some combination of all bio-
chemical reaction rates in a cascade of reactions. The rate kwill there-
Nmax
1 N
N
m
0ax(^1) ekt