Predicting Fermentation Rates in Ale, Lager and Whisky

Recently there has been an increased interest in characterising the rates of alcoholic fermentations. Sigmoidal models have used to predict changes such as rate density decline. In this study, three published sigmoidal were assessed and fit industrial fermentation data. The first is four-parameter logistic model described ASBC Yeast-14 method. second a nested form function, adding extra parameter, creating 5-parameter equation., where additional parameter was added allow for asymmetry. final three-parameter equation which describes change Apparent Degree Fermentation with time. compared by fitting them data from Australian Canadian lagers, American Scottish ales Scotch Whisky fits then one another technique developed Akaike F-test. information criterion compares accounts both goodness fit, number parameters model. Finally, consideration given establishment prediction bands (that enclose area that can be 99% sure contains true datapoints). Calculation these “challenging” but successful heteroscedastic.