The implementation of fuel cells (FC) in transportation systems such as airplanes requires better understanding of their mechanical behaviour in vibrating environment. To this end, a FC stack was tested on a vibrating platform for all three orthogonal axes. The experimental procedure is described in the first part of the paper. This second part of the paper demonstrates how the experimental data collected can be used to create a three-dimensional, multi-input and multi-output model based on the Artificial Neural Network (ANN) approach. Indeed FCs are nonlinear mechanical systems, difficult to be physically modelled. The ANN methodology which depends strictly on raw data is a particularly interesting alternative solution to model FCs, for example, for monitoring purpose. The ANN model is described along with the training, pruning and validation stages. The results are exposed and commented. (C) 2008 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.