3.2. Exposure assessment
In this study the exposure assessment refers only to the ingested number of cells of SFP by the consumer, though exposure assessment can include also microbial toxins and toxic chemicals. A food SC can be very complex and can have many steps and processes. According to the Modular Process Risk Model (MPRM) (European Commission – Health, 2003 and Nauta, 2002), it is assumed that at various stages of the SC one of the six basic processes can be assigned, i.e., (i) growth, (ii) inactivation, (iii) partitioning, (iv) mixing, (v) removal and (vi) cross contamination. The input parameters in each of these modules are the number of the microorganisms and their prevalence. The number of the microorganisms and their prevalence for each successive process could be obtained in two different ways: (i) by using the mathematical model to estimate the number of the microorganisms and their prevalence at the end of the previous process, which automatically become the input variables for the next process in the chain, or (ii) by measurement in control points between processes.
Each step in a SC presented in Fig. 2 could be classified in one of the six basic processes related to the various stages in the SC. Basic input and output variables in each step are the concentration of bacterial organisms (N) and prevalence (P). These quantities should be considered as uncertain variables in the model. The variability in the model is a consequence of the variability in the environmental conditions as well as variability in the SC. The variability in the SC includes uncertainty in different steps of the SC, such as exact time when the product arrives or departs certain point along the SC. The uncertainty in different environmental and SC parameters can be included into the mathematical model by using stochastic models, e.g., Monte Carlo simulation. The MPRM is not straightforward to implement and requires significant amount of data which is not usually available. Further in this paper only microbial growth and cross contamination are considered without reducing the generality of the conclusions in this work.