Aiming at the problem of aircraft conflict prediction and resolution strategy in continuous descent operation （CDO）， a four-dimensional trajectory （4DT） prediction model is constructed to realize the accurate prediction of flight conflicts. We take fuel consumption and conflict duration as the optimization objectives， and aircraft speed and descent path angle （DPA） as the optimization variables. Thus， flight conflict resolution is realized based on a multi-objective genetic algorithm， NSGA-Ⅱ. Finally， a multi aircraft CDO flight in a terminal area is taken as an example for conflict prediction and resolution， and the influence of the optimization target weights on the average fuel consumption and the optimization variables in airspace is analyzed. The results show that the proposed algorithm can realize multi-aircraft conflict-free CDO in the terminal airspace. Compared with the values before optimization， the average fuel consumption of the 20 aircraft is saved by 11 kg/fight， and the cumulative flight conflict time between aircraft in airspace is reduced from 984 s to 0， which eliminates the flight conflict. The research results are helpful to realize multi-aircraft conflict-free CDO in airspace， and can improve the implementation rate and operation effect of CDO in busy airports.