Student achievement is a national and institutional phenomenon, studies have found that engineering programs are particularly vulnerable, with national throughput rate of 17% after 5years of study. The severity of the problem has led to the exploration of techniques which can be used to predict student performance after access to higher education.

Artificial Neural Network (ANN) and linear regression a model was used to predict the total number of subjects passed in semester 1 of 2008. Data of the Tshwane University of Technology was used for this study. The total Average Point Scores (APS) students got in grade 12 was used as input variable. The results indicate a better agreement between ANN model prediction and observed values compared to the linear regression. It demonstrates that the ANN-based model developed can predict the total number of subjects passed in semester 1 with high accuracy.