Data-driven models of roller mill and plansifter operations of a wheat milling process have been developed.

The roller mill models are used to improve process understanding and to predict the particle size distribution (PSD) of the milled product.

The plansifter models are used to predict the PSD and the amount of material on each sieve.

Very satisfactory prediction of the PSD and sieved material amount is obtained.

A comprehension of the wheat kernel breakage mechanism is obtained using a single modeling framework that unifies the modeling results obtained in existing studies.