Assessment of water resources system resilience under hazardous events using system dynamic approach and artificial neural networks

Abstract The objective of this research is to propose a novel framework for assessing the consequences hazardous events on water resources system using dynamic resilience. Two types were considered: severe flood event and an earthquake. Given that one or both hazards have occurred considering intensity those events, main characteristics resilience evaluated. utilizes artificial neural network (ANN) estimate ANN was trained large, generated dataset included wide range situations, from relatively mild ones. A case study performed Pirot (Serbia). Dynamic derived developed dynamics model alongside models implemented. most extreme hazard combination results in robustness 0.04, indicating earthquake with significant magnitude hydrograph low frequency occurrence. In moderate hazards, has median value 0.2 rapidity 162 h. ANN's efficacy quantified average relative error metric which equals 2.14% 1.77% rapidity, respectively.