A real method of predication brake pad wear ,could lead to substantiol economies of time and money. This paper describes how such a procedure has been used and gives the results to establish is reliability by comparing the predicted wear with that which actually occurs in an existing service. The experimental work was carried out on three different commercial samples ,tested under different operation conditions (speed,load,time...etc)using a test ring especially modified for this purpose. Abrasive wear is mainly studied , since it is the type of wear that takes place in such arrangements. Samples wear tested in presences of sand or mud between the mating surfaces under different operational conditions of speed, load and braking time .Mechanical properties of the pad material samples (hardness, young,s modulus and collapse load under pure bending condition )wear established . The thermal conductivity and surface roughness of the pad material wear also found in order to enable comparison between the surface condition before and after testing. Sliding velocity had a small effect on the wear rate but it had great effect on friction coefficient. Wear rate was affected mainly by the surface temperature which causing a reduction friction coefficient and increasing the wear rate. Surface roughness had almost no effect on the wear rate since it was proved experimentally ,that the surface becomes softer during operation .mechanical properties of the pad material had fluctuating effect on wear rate. The existence of solid particles between pad and disc increasing wear rate and friction coefficient while the mud caused a reduction in wear rate of the pad surface since it acts as a lubricant absorbing the surface heat generated during sliding the area of contact between pad and disc. wear rate obtained experimentally agreed fairly well that found from empirically obtained equations.