Statistical Derivation of Sampling Plans for Microbiological Testing of Foods

Sampling food for microbial testing is a risk management strategy used to evaluate whether a food safety system is correctly implemented. Within-batch testing can be accomplished to comply with microbiological criteria or to assess whether the food production process is under control. Generally, sampling plans can be designed to meet the consumer’s and/or producer’s quality requirements, and have been conventionally derived by borrowing notions of acceptance sampling theory from classical quality statistics. This chapter describes and illustrates the methodologies to derive the different types of sampling plans used for microbiological criteria in foods; namely, the two-class attributes sampling plans (based on prevalence, on concentrations, and with an enrichment step) and the variables sampling plans. The chapter also discusses the weakness of the classical assumption that the measure of the quality characteristic (i.e., log microbial concentration) is normally distributed among food units with a variance that is approximately stable batch to batch; and provides an insight to novel modelling trends to produce more efficient and discriminatory sampling plans.Dr. Gonzales-Barron wishes to acknowledge the financial support provided by the Portuguese Foundation for Science and Technology (FCT) through the award of a five-year Investigator Fellowship (IF) in the mode of Development Grants (IF/00570).