The relevancy of dynamic simulations is investigated through the modelling of two process safety systems in Aspen HYSYS Dynamics V9. On the one hand, the design of flare systems is conventionally made according to steady-state simulations, which even though is conservative, in turn leads to oversizing of the system beyond feasibility. API Standard 521 suggests that an improvement in the design can be achieved by using dynamic models to account the dynamic nature of the operation. The emergency depressurization of the flare system of three different facilities (in terms of system size, flare design rate, network pipe dimensions, and total hold-up volume) have been modelled, and results have been benchmarked against the steady-state design. It is found that the larger the flare system, the larger the hidden potential for debottlenecking, i.e. larger difference between the steady-state design rate and the dynamic peak flare rate. On the other hand, the process response time available for the valve closure of the High Integrity Pressure Protection System (HIPPS) is analysed, and the relation between robustness of the model and sensitiveness of the results is investigated. It is observed that a more conservative prediction can be achieved when using the pipe flow correlation Tulsa Unified Model.