Dough fermentation is a fundamental step in bread manufacturing, commonly supervised by bakery experts. This work aims to develop an efficient technology based on near infrared spectroscopy (NIR) and partial least squares-discriminant analysis (PLS-DA) to determine the fermentation state of bread loaves in the bakery industry. Knowing the fermentation state of the loaves during the manufacturing will allow to act in the production line, avoiding introducing an inadequate loaf in the oven and reducing costs. For that, a new methodology is proposed that consists in reproducing the knowledge of a Master Baker through a NIR spectrometer. Regarding the sequentiality of the objects and the real applicability of the method, three different cases were proposed using PLS-DA, getting that, in the best case, the sensitivity of the prediction set for the unfermented doughs was 88% and for the fermented and over-fermented was 86%; whereas the specificities were all greater than 86%.
