This paper investigated the performance of a regenerative heat and mass exchanger (RHMX) for indirect evaporative cooling. A numerical model for RHMX was developed and validated with experimental data from the literature. Then, a data-driven model based on artificial neural" network (ANN) algorithm was presented which was derived from the simulation results generated from the numerical model. The comparison between ANN prediction and experiment data showed good prediction accuracy. The average prediction error between the predicted and tested data was around 4% based on the air temperature change across the dry channel. With the data-driven model, parametric analyses were made to investigate the performance of the RHMX under different operating conditions. Finally, a design optimisation of the extraction air ratio was conducted under different ambient conditions. It was found that the optimal extraction air ratio decreased with the ambient temperature and/or relative humidity which ranged from 0.3 to 0.36. (C) 2017 Elsevier B.V. All rights reserved.