In the present research work predicted properties of Aluminium Alloy Composites prepared through powder metallurgy technique have been examined using Artificial Neural Network (ANN) approach. The aluminium alloy (Al-20Fe −5Cr) matrix reinforced with aluminium oxide (Al 2 O 3 ) varying from 0-30 wt% with a step of 10 wt% were prepared. The green compacts prepared at three different compaction pressures viz. 470, 550 and 600 MPa were sintered at 440 °C for 30 min. Pin-on-disc test was performed to evaluate wear loss of the composites prepared. The influence of alumina varying weight percentage and different compaction pressure had been analyzed and tests were performed according to full factorial design. Analysis of variance (ANOVA) had been employed to predict the percentage contribution of various process parameters. The results indicated that the wear loss was mainly influenced by alumina percentage followed by compaction pressure by 92% and 8% respectively. For modeling and prediction of wear loss, a feed forward back propagation neural network was constructed and compared with experimentally calculated values. Both experimental and ANN predicted values of wear loss were in close correlation with each other with approximately (1-3) % error. RECEIVED