The Aluminuim metal matrix composites [MMC's] have been widely sought over conventional materials for marine, aerospace and automotive applications owing to their excellent wear resistance and mechanical properties. Al7075 alloy though possess high strength and toughness, they exhibit poor resistance against wear and to enhance this property the matrix has been reinforced with ceramic particulates of Al2O3. The MMC was fabricated using vacuum assisted stir casting technique by reinforcing the matrix with Alumina particulates of average 150μm with weight percentage varying from 2%-6%. Wear results in damage of machine components and in order to analyze this complex phenomenon, experimentation was carried out using L27 orthogonal array technique according to ASTM G99 standards. The loss of material in terms of wear-height of the base alloy Al7075 and Al2O3 reinforced composites increased with load applied and sliding distance, similarly trend was observed to decline with an increase of the reinforcement Al2O3 in the matrix material. An Artificial Neural network[ANN] was established for the forecast of tribological properties of the Al7075-Al2O3 composites using Levenberg-Marquardt optimization technique to establish a nonlinear relationship between density, wear height loss, sliding distance and weight percentage of particulate reinforcement. A good agreement has been witnessed between experimental and ANN model results. After ANN prediction confirmation tests were carried out for experimental results for verification. © 2017 Elsevier Ltd.