We apply the Seismic Network Evaluation through Simulation (SNES) method to evaluate the performance of the Alaska Regional Seismic Network (ARSN) in locating regional earthquakes in terms of validity of the velocity models, hypocentral errors, and magnitude of completeness. We find that background noise levels at the sites are highest in the Aleutian region and lowest in mainland Alaska. We demonstrate that the ARSN provides the best monitoring in the south‐central region, with horizontal and vertical errors of less than 1.5 and 3 km, respectively, at the completeness level of ML 2.0. At the completeness level of ML 3.0, the network is capable of locating earthquakes for the whole region down to depths of about 250 km with satisfactory errors. The lowest magnitude completeness levels, of about 1.4 and even less, are achieved with satisfactory location errors in limited areas of the south‐central region that have the densest station coverage. The network does not monitor the seismogenic areas in northern, western, southeastern, and offshore Alaska at an adequate level in terms of earthquake location errors. Network upgrades in these areas could be optimized using the SNES results.