Flow behaviour of powders plays an important role in defining their performance in many industries. In this study, we present our work on prediction of flow characteristics of binary and ternary mixtures under dynamic conditions using Discrete Element Method and simulating the expended mechanical work of a rotating impeller penetrating a packed bed. Three commonly used methods for calculating a mean have been explored to express the mixture Bond number, namely arithmetic, geometric and harmonic mean. This is done by introducing a weighting factor based on the fractional surface area of each component of the mixture. The mixture Bond number is dependent on the interfacial surface energy, particle size and density; and a wide range of Bond number is covered in our study by varying all three. The expended work of impeller shows the clearest trend with the mixture Bond number when it is expressed in terms of the arithmetic mean. Although we only used binary and ternary mixtures (40 different mixtures in total) in this study, the trend should be applicable to multi-component mixtures and therefore useful as a design aide for powder formulation. (C) 2020 Published by Elsevier B.V.