The study of abrasive wear resistance of alloys used for high mechanical and thermal stress applications, particularly those used in advanced stages of jet engine compressors, involves a thorough theoretical understanding of the behaviour, as well as a robust experimental and computational analysis. The creep-resistant Ti8Al1Mo1V (Ti811) is used in compressor stages, subject to substantial amounts of abrasive wear. Based on an experimental analysis of the abrasive wear resistance of the heat-treated alloy, which involved a pin-on-disc wear test regime according to the ASTM G 99 standard, two input parameters (sliding velocity and normal load) were arrived at and were used to determine the values of the rate of wear and the coefficient of friction at low, moderate and high input values. The results of these were used as training data for four predictive models Random Forest (RF), Support Vector Regressor, XGBoost, Ensemble of RF &XGBoost) and performance of each of these was tallied against various output parameters. Judging by the performance metrics of Mean Square Error, Goodness of Fit Measure (R2) and Mean Absolute Error, the ensemble (RF + XGBoost) model proved to be the best option for predictive modeling among the four.