Food safety risk assessment for the Australian dairy industry

This thesis describes the development of a food safety risk assessment framework for the Australian dairy industry, through the collection of specific information regarding that industry and its collation and organisation into a structured and flexible spreadsheet model. The risk assessment model framework was developed for use as a tool for evaluating, identifying and prioritising research needs. Developed in Excel (Microsoft Corp.) with @Risk (Palisade Corp.) as an add-in, the model uses a stochastic approach to evaluate the likely concentration of hazards that may be present in liquid milk. Those hazards include: Escherichia coli, Listeria monocytogenes, Salmonella spp., Campylobacter spp., Staphylococcus aureus and Yersinia enterocolitica, Bacillus cereus, antibiotic and herbicide residues. The structure of the model allows for multiple hazards to be modelled in a single simulation. Flexibility of the spreadsheet model allows for manipulation of the input distribution 'parameters. This enables the evaluation of the effect on hazard concentration in liquid milk to be determined from specific. contamination events at the farm level. The effect of pasteurisation on microbiological hazards was examined together with the validity of log-linear thermal inactivation kinetics. Model estimates for antibiotic residues show that there is less than 10-3 chance of exceeding the MRL when starting with an initial probability of 0.0016 that each cow is contaminated and 0.1 grams of residue is present in milk from each contaminated cow. Microbiological hazards are not expected to survive pasteurisation given the estimated concentrations of those hazards entering the milk supply from the farm level and the reliability of the pasteurisation process.