The performance of the machining (turning) process is evaluated in terms of tool life, surface roughness, tool-shim interface temperature developed and metal removal rate during the process. It is very important for the manufacturing engineers to know the performance of the turning process for a set of cutting (input) parameters. In this paper, parametric models based on multiple regression analysis (MRA), neural networks (NNs) and case-based reasoning (CBR) are developed for predicting the machining performance, i.e. the output parameters. An experimental database containing 114 data sets are used for developing the three models. Each data set contains nine input and four output parameters. About 20 machining trials are exclusively conducted with various combinations of input parameters, and their corresponding output values are compared with the predicted values of the developed models. Descriptive statistics of the errors are calculated for the three models and it was found that the CBR model provided better prediction capability than MRA and NN models. © 2012 Inderscience Enterprises Ltd.