Batch dispersion model to optimise traceability in food industry

Facing many food safety crises, like BSE or foot-and-mouth disease, food companies try to limit incurred risk and to reassure consumers. So today, the point is not only to trace the products efficiently but also to minimize recalls and the number of batches constituting a given finished product. The problem studied concerns a sausage manufacturing process in a French food company. It tries to minimize the quantity of recalls when products are characterized by a 3-level ‘disassembling and assembling’ bill of material. Such a ‘dispersion problem’, encountered in the food industry, has been modelled, solved and experimented. A mathematical MILP model is proposed and the results of experiments obtained with LINGO software are presented.