The intrinsic variability of the grain, the equipment layout characteristics, and the intricacy involved in the flour milling process flow make it very complex to analyse and to estimate the impacts associated with alterations in a particular stage of the flour milling system. Computer simulation is a desirable tool that can aid in the understanding and analysis of the performance of a processing facility, especially a flour mill. In this study, computer simulation models were developed to facilitate the analysis of quality characteristics of the mill streams. The quality characteristics studied were moisture, ash and protein. Using probabilistic functions obtained from experimental data in the Kansas State University Pilot Mill, three models were developed to simulate the moisture, ash and protein flows. The models had an excellent performance in the verification process showing none, one and three streams outside the control limits analyses (99.99% confidence interval) for the moisture, ash and protein models, respectively, from a total of 126 streams. Using the same control limits analyses in the validation process, the difference between results from the system and results from the models increased to three, three and four streams for the moisture, ash and protein models, respectively. The control limit analysis indicated that the variation was related more to the fluctuations of the system rather than to the models per se. Four different scenarios were simulated with the models and the results were analysed. These models present a great potential in industry applications to evaluate the effects of new raw materials, equipment modifications, and changes in processing conditions on the milling process and its products.
