The prediction of the wear rates and coefficient of friction of composite materials is relatively complex using mathematical models due to the effect of the manufacturing process on the wear properties of the composite. Therefore, this work presents a rapid reliable tool based on neural network modified with particle swarm optimizer to predict the wear rates and coefficient of friction of Al-TiO2 nanocomposite manufactured using accumulative roll bonding (ARB). The wear rates and coefficient of the produced composites were computed using pin-on-disc and correlated with the composite morphology, hardness and microstructure. Experimentally, it was demonstrated that the hardness and wear rates reduce with increasing the number of ARB passes until a plateau was achieved due to the uniform distribution of TiO2 nanoparticles inside the composite and the saturation of grain refinement in the Al matrix. The maximum hardness improvement was 153.7% for composite containing 3% TiO2 nanoparticles after 5 ARB passes. While the wear rates of the same composite tested at 5 N load reduces from 3.7 × 10−3 g/m for pure Al to 1.1 × 10−3 g/m. The proposed model was able to predict the wear rates and coefficient of friction for all the produced composites tested at four different wear loads with excellent accuracy reaching R2 equal 0.9766 and 0.9866 for the wear rates and coefficient of friction, respectively.