Engineering adequate disinfection processes and assessing risks associated with various disinfection options requires knowledge of kinetics of microbial inactivation as a function of design variables (e.g., dose, contact time). Often such information is obtained in batch studies and extrapolated to design conditions. By Monte Carlo techniques, we have shown that the use of a direct maximum likelihood evaluation for two types of data normally encountered, counts (PFU or CFU) and dilution experiments (MPN), leads to estimates of microbial inactivation rate parameters having lower bias and variance than other data reduction normally employed. This paper reviews the methodology of this technique and provides the theory for the computation of confidence limits for parameter estimates, and discusses how the data may be checked for consistency (goodness of fit determination).

This content is only available as a PDF.
You do not currently have access to this content.