Grinding of rails is a technique widely used within the railway industry to balance the degradation of the condition of the rail with the required performance of the rail. The principal focus of this research is the impact of wear and rolling contact fatigue (RCF) cracks on structural integrity of rails, and how rail grinding affects this relationship. A numerical model which predicts growth of RCF-initiated cracks in rails has been adapted to take into account periodic grinding of the surface of the rail. The suitability of some of the simplifying assumptions of the adapted model, referred to as the Grinding Model, has been examined with a physical test program, using full scale rail vehicles and track. This test program studied the persistence of the characteristic surface roughness of the rail generated by grinding, and was carried out to determine whether the effect of this roughness on crack growth can be neglected in the Grinding Model. The Grinding Model has been used to predict crack size, in order to investigate the effect of different grinding strategies, consisting of a depth of grinding applied at a certain interval during a representative pattern of rail vehicle traffic over the rail. The use of the Grinding Model to find grinding strategies which match an optimum criterion has been demonstrated. The applicability of this optimisation technique and the model in its current state of development, to the specification of rail grinding operations, in the context of maximising safe rail life and minimising rail life cycle cost, is discussed.