Electrical discharge machining of titanium alloys requires a balanced selection of machining parameters, as high discharge intensity can improve material removal while increasing electrode wear, overcut (OC), energy consumption, and surface damage. This work investigates the sustainable EDM of Ti-6Al-4V using deionized water as the dielectric medium and copper/brass electrodes. Peak current , pulse-on time , and gap voltage were varied using a central composite design, and the responses were evaluated as material removal rate (MRR, m ), tool wear rate (TWR, ), OC ( µ m), and discharge energy (J ). MRR was calculated from workpiece mass loss and titanium alloy density, while TWR was calculated from electrode mass loss, electrode density, and machining time. Response surface methodology was used for factor screening, and artificial neural network modeling was used to predict nonlinear machining responses. Particle swarm optimization was then applied, using the trained ANN as the fitness evaluator, with TLBO as the benchmark optimizer. The optimized copper-electrode condition was A, µ s, and V, producing MRR of 6.1 , TWR of 0.003 , and surface roughness of 1.4 µ m. SEM and wear-track analysis confirmed the need to include surface integrity and tribological response during EDM parameter selection. Future innovation will extend the quantitative range of recast-layer thickness and surface chemistry, and evaluate real-time discharge monitoring for scalable industrial EDM control.