Statistical and computational intelligence techniques were employed for informatics based design of nano friction modifiers added bio-lubricant. Systematic data were generated through laboratory experiments, using design of experiment, to study the effect of addition of multi-wall carbon nanotubes as friction modifiers in castor oil on frictional properties. The experimental data were used to develop data driven models using statistical techniques, artificial neural network and fuzzy inference systems. The simulation studies which were based on the model predictions were used to design the nano-lubricant with multi-walled carbon nanotubes as the friction modifiers. The optimum combination of nanotube concentration and load, found from the model predictions, were experimentally validated. To understand the lubrication mechanisms, the surfaces of the pin which were tested with the optimum concentration of nanotube added castor oil were characterized with scanning electron microscope and energy dispersive systems, which showed the adherence of friction modifiers on the steel surface.