The present study aims to investigate the use of Artificial Neural Networks (ANN) for the performance-based condition monitoring of indusrial gas turbine engines. Toward this end, a health assessment tool is presented by developing a Multi-Nets ANN model. A number of key performance parameters that are commonly measurable on the most industrial gas turbines are monitored and their associated neural networks for the healthy condition are trained. Three-layer feed-forward configuaration is chosen to construct the networks, the Levenberg-Marquardt algorithm is used as the training function, and the k-fold cross-validation process is employed to obtain the optimum number of neurons in the hidden layers. The model is developed and tested using the gas path performance data collected from an 18.7 MW twin-shaft industrial gas turbine. A special attention is also devoted to the system theory interpretation in order to evaluate the effect of the input neurons on each output of the Multi-Nets. To that end, the sensitivity analysis is formulated using derivatives based on an interpretation of the neural network’s weights.