Assessing the optimal experiment setup for first order kinetic studies by Monte Carlo analysis

In inactivation studies of microorganisms and quality influencing enzymes a log linear relation between dependent and independent variables, generally denominated as a first order kinetic, is frequently encountered. Reliable application of a kinetic model to predict inactivation requires a proper quantification of the variation on the model parameters. The aim of the present research is the assessment of the most optimal experiment setup leading to first order kinetic parameters with minimal variation, and, by consequence, to model predictions with minimal variation. As a vehicle for this research, the first order inactivation of pectin methyl esterase (PME), commonly encountered in fruits, is considered. Based on a bootstrap assessment of the PME activity measurement variation, a Monte Carlo analysis fully reveals the optimal experiment setup and leads to two important conclusions, valid for all first order kinetic studies. First, if the logarithm of the dependent variable has a constant variance as function of the independent variable, the optimal sampling scheme is a 50–50 division at the two extremes of the independent variable range. It is indicated how this relates with classical linear regression analysis. Second, if the logarithm of the dependent variable has a non-constant variance, this variance should be fully characterized and the optimal sampling scheme should be obtained via Monte Carlo analysis. It is shown how, in such a case, a 50–50 division is not necessarily the most optimal.