Kushner et al. proposed a scale-up model to describe the effect of lubrication on tablet tensile strength. In our study, a lubricant cohesiveness term, (1 + alpha/V), was added to the lubrication scale-up model reported by Kushner et al. to address the cohesiveness of magnesium stearate. The modified model proposed is as follows: TSSF=0.85 = TSSF=0.85,0[beta(1 + alpha/V)exp(-gamma x 1.5 V-1/3 x F-headspace x r) + (1 - beta)], where TSSF=0.85 is tablet tensile strength at solid fraction 0.85, V is blender size, F-headspace is fraction of headspace, r is revolution number, TSSF= 0.85,0 is theoretical initial tensile strength, alpha is the sensitivity constant of magnesium stearate powder cohesiveness, beta is the sensitivity constant of the blend to lubrication, and gamma is the lubrication rate constant. Using a high shear placebo granule, lubrication scale-up models were built using both equations, with and without the cohesiveness term, and tested by a 16-fold scale-up. The equation with the cohesiveness term demonstrated superior prediction of tablet tensile strength for scale-up within the selected granular system and the scale-up range studied in this work, indicating the usefulness of including magnesium stearate powder cohesiveness in building a scale-up predictive lubrication model using an active pharmaceutical ingredient (API) sparing approach with small quantity of materials. (C) 2020 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder Technology Japan. All rights reserved.