The　increasing　demand　for　mitigating　earthquake　hazards　has　prompted　substantial　research　attention　towards　performance-based　seismic　design　of　civil　structures.　Nevertheless,　there　remains　limited　exploration　into　optimizing　complex　structures　while　accounting　for　seismic　uncertainties.　This　study　seeks　to　address　this　gap　by　introducing　an　effective　approach　for　optimizing　designs　of　nonlinear　structures　under　random　seismic　excitations.　The　key　innovation　lies　in　approximating　structural　failure　probability　through　incremental　dynamic　analysis　(IDA),　leading　to　the　development　of　a　novel　double-loop　optimization　method　tailored　for　designing　nonlinear　structures　exposed　to　stochastic　seismic　loading　conditions.　In　the　outer　loop,　geometric　variables　of　structures　are　optimized　using　sequential　quadratic　programming;　within　the　inner　loop,　IDA　is　adopted　for　structural　analysis　to　quantify　seismic　uncertainty,　and　the　resulting　failure　probability　is　then　served　as　the　optimization　constraint　for　the　outer　loop.　To　validate　its　accuracy　and　efficacy,　numerical　investigations　have　been　performed　on　two　representative　case　studies　utilizing　OpenSees:　a　reinforced　concrete　column　and　a　threestory　steel　frame.　The　findings　affirm　that　IDA　can　precisely　estimate　failure　probabilities　associated　with　nonlinear　structures　experiencing　random　ground　motions　and　demonstrate　that　this　proposed　methodology　can　effectively　determine　optimal　geometries　aimed　at　enhancing　structural　resilience　against　earthquakes　across　various　levels　of　failure　probabilities　and　bound　constraints.