Acrylamide formation in a food model during frying can be analyzed in space and time with a traditional heat and mass transfer approach. In this paper a joint simulation/experiment approach has been exploited: the computational fluid dynamics has been coupled to chemical kinetics, to describe concentration and temperature fields in a food substrate, subject to interdependent and non-uniform heat and mass transfer, while validating measurements have been performed by high performance liquid chromatography. A multi-objective SIMPLEX optimization has been finally employed for the overall model tuning, incorporating all phenomenological variations. The numerical results confirmed that the acrylamide formation is non-linearly dependent on the operating thermal regime, in general more than doubling its average levels when the process temperature is increased by 10% only. Based on the proposed acrylamide maps, a maximum concentration locus is found on the product's upper corner, depending on the specific transport ensemble and therefore on the local process conditions. For potato frying at an initial temperature of 190 degrees C, the maximum local acrylamide level after 4 min (almost 4 x 10(-2) mg/g) is almost four times the one obtained when frying at 170 degrees C, while scaling only linearly with process duration. (C) 2009 Elsevier Ltd. All rights reserved.