Nowadays technology frequently makes use of IT components for implementing communication between the various elements that form a system. CPS (Cyber Physical Systems) are increasingly present in the industrial world and their application is transversal in all sectors. While this allows for greater efficiency, coordination and quality, it also opens to a series of completely new disruption scenarios with a potentially disastrous impact. In this context, the analysis of the system security and reliability pushes for a novel cyber-resilience perspective capable of evaluating not only the possibility of failures related to the physical and tangible part but also anomalies of and attacks to the IT counterpart. This work investigates this research dimension with respect to a water purification and distribution system. The presence of multiple documented events regarding cyber-attacks against the water industry and the consequences that such accidents could have on public health, further motivates this choice. A water desalination plant model and three cyber-attack scenarios have been developed in MATLAB/Simulink. To evaluate the societal impact of plant failures, a georeferenced water distribution network has been designed on EPANET. A set of resilience metrics has been customized for the system at hand, following a probabilistic approach that take into consideration the variability of cyber-attack’s duration and system response, along with the physical modeling of the process. The results provide evidence on the benefits of cyber-physical inspired simulation for understanding the behavior of a system, identifying its criticalities, and designing corrective actions.