Evaluation of Novel Quality Control Methods for Wheat Flour

Abstract: The quality of a flour can vary greatly depending on the wheat variety and on the conditions while growing the wheat. Currently, the quality is mainly assessed by test baking the flour to observe the volume of the final product. The aim of this study was to investigate if three instruments (Alveolab, Mixolab and Rheo F4) analyzing rheological behaviours connected to dough preparation and baking properties, can provide additional valuable information to current flour quality controls. To achieve this, Mixolab and Rheo F4 analyses were conducted on a large set of flour samples, and the results were processed with the multivariate analysis methods PCA and PLS. Ultimately, predictive models for baking volume were created and evaluated. Furthermore, different preparation methods of doughs were tested in Mixolab and Alveolab to investigate how well these methods translate to the industrial bakery process. The results from the study indicate that the examined instruments alone are not able to give an accurate prediction of the baking volume. However, they seemed to provide a minor positive impact when creating a model where other quality control parameters and parameters connected to the chemical composition were included. This model obtained a predictive power, Q2, of 0.834, suggesting that additional information is needed in order to create a predictive model that is valid for industrial use. It was further seen that the standard protocols in Mixolab and Alveolab provide results vastly different from those obtained when preparing the analyzed doughs according to industry standards. To conclude, the methods do not seem to give enough additional valuable insights on the final quality to justify the extra cost and time it would take to include it in quality controls.