A mapping tool developed by government scientists to aid adversarial risk assessment—predicting areas of the food chain where malicious contamination resulting from sabotage might occur—could also assist public health officials and industry in pinpointing the sources of contamination when an accidental outbreak of foodborne illness occurs.
“Stochastic mapping” shows what is known about how products flow through the distribution supply chain and characterizes the uncertainties in supplier-customer relationships that result from incomplete information.
“Imagine a food supply network where you know all the connections; you’re a food processor, and you know all the suppliers and all the customers,” said Steven Conrad, a systems analyst in the Infrastructure Modeling and Analysis Department at Sandia National Laboratories in Albuquerque, N.M. “If someone’s getting sick from peanut butter, it would be pretty straightforward to walk it back to the store they bought it from and the distributor the store got it from, and so on.
“But real life is not like that. There are always uncertainties about who sold what to whom. With our probability mapping, we build all that in; where we’re sure there are relationships, we assign a probability of one, and where there are no relationships, there’s a probability of zero.”
When the degree of interaction between two entities in the food chain is uncertain, scientists assign a probability number based on the degree of likelihood of their interaction. In the event of an outbreak, such a map could be used to suggest the most likely source of the contamination and—given limited resources and time constraints—the one to investigate first.
The model proved effective in a study published recently in the International Journal of Critical Infrastructures in which Conrad and his colleagues tested it in a proof-of-concept analysis involving the fresh sprout sector in one region.