Data from an industrial case study of breakfast cereal production indicated that the generated amounts of acrylamide are greatly dependent upon the combined effects of temperature and heating time in a roasting step process. Two approaches to obtain process models for acrylamide generation were tested. The first applied a pathway-based model. The second developed a simpler more robust model based on the integrated effects of time and temperature, where the generation of acrylamide was crudely fitted to an exponentially rising function. The development of the two models highlighted a number of difficulties in applying multi-parameter models and emphasized the advantages of "classical" approaches to process modelling, especially for use in an industrial context. The study faced with a significant degree of variability in the data, due to fluctuations in the process, which also emphasized the importance of robustness in the developed models. The correlations obtained for predicting acrylamide generation in the case study present a useful tool for food processing industry to minimize acrylamide generation. In the present case it was possible by lowering process temperature and prolonging residence time to achieve an approximately 80% reduction in acrylamide content while maintaining the desired product quality. (C) 2007 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.