| Abstract: | Reliability analysis is especially important when critical decisions are to be made involving potentially severe adverse consequences such as foodborne illness. Owing to uncertainty associated with the parameters controlling survival of Listeria monocytogenes in chorizo (a Mexican‐style sausage), the time needed to reduce the count by a certain number (n) of logs (tnD) is probabilistic. In this paper the first‐order second‐moment (FOSM) method based on Taylor series expansion is used to derive the expected value and standard deviation of tnD as function of the operating conditions (random variables) affecting survival, namely initial water activity (aw0) of the sausage batter, storage temperature (T) and airflow velocity (F), along with their uncertainties characterised by their means and coefficients of variation. For any given n the derived tnD probability distribution enables one to determine an estimate of tnD for any desired level of reliability or confidence level, such as 50% (median value), 95%, 99%, etc. Among the conclusions drawn were: (i) the variability associated with T and F has a minor effect on estimating uncertainty in tnD, whereas the reliability of tnD estimation is greatly influenced by the uncertainty in aw0; and (ii) the uncertainty in aw0 has the greatest impact when aw0 of the sausage formulation exceeds 0.90. The approach used and discussed in this paper can be applied to any survival/inactivation study to incorporate the effect of uncertainty in the various extrinsic and intrinsic parameters on the survival kinetics of the pathogen in a food system under evaluation. Copyright © 2006 Society of Chemical Industry |
