This paper presents a novel global/local optimization approach for finite element model updating of a composite flying-wing aircraft that used subcomponent-based test data. Three-steps are considered to update the mass and stiffness distributions for use in the finite element model updating of the composite flying-wing aircraft. Steps I and II, i.e., local optimizations, update the mass distribution for the centerbody and the individual wings of the aircraft, respectively, to match their mass properties with the test data. The individual wing stiffness distribution is also updated in Step II by using available wing’s ground vibration test results. The updated finite element models for the centerbody and the wings are then assembled as the initial finite element model for the full aircraft model, in Step III, i.e., global optimization. This initial finite element model is then updated using available experimental mass properties and ground vibration test modal results for the full aircraft. The global/local optimization iterations continue till the differences between the test data and numerical results for models of both full aircraft and subcomponents are less than given criteria. The proposed approach on finite element model updating is applied for a composite flying-wing aircraft, mAEWing2, used in the NASA Performance Adaptive Aeroelastic Wing program. Results show that the mass and modal results for the updated finite element models of the components and the full model agree well with the test results. The updated finite element model is also verified by comparing its static deformations and tip accelerations frequency response function with the test results. A good agreement is observed between the finite element analysis and test results.