This work provides a large-scale trajectory optimization approach based on air traffic complexity to balance the overall airspace situation under the trajectory-based operation mode. It uses real operation data simulation to verify its effectiveness and optimization effect. Firstly

an air traffic complexity calculation model is constructed based on the potential interaction relationship between flights. Secondly

a multi-objective large-scale trajectory optimization model that meets the operational requirements of air traffic control is constructed based on the air traffic complexity calculation model

and a high-quality genetic solution algorithm is proposed. Finally

using the national flight operation data from June 2019

a simulation simulation of air traffic complexity-based trajectory optimization is performed

and a comparison between conflict-free trajectory optimization and air traffic complexity-based trajectory optimization is conducted. Conflict trajectory optimization is compared and analyzed. The simulation results show that the proposed method can resolve 93.74% of potential conflicts. Compared with conflict-free trajectory optimization

its optimization scheme exhibits less air traffic complexity fluctuations when facing environmental perturbations such as waypoint waiting and area bans. By adjusting 21.11% of the flights

it can reduce the average complexity of each time period by 24.98% on average

and the overall average complexity of the whole day is reduced from 120.52 to 72.82.