Collaborative management of logistics and assembly processes plays a significant role in fit-out construction projects. Facilitated by digital twins, multi-dimensional information about fit-out resources and operations are updated to generate the corresponding digital twins to support logistics-assembly synchronization. However, the information integration and updating in such a multi-source and heterogeneous data environment are fragmented, and coordinated decisions that coordinate the resources and operations in a real-timely synchronized manner is hard to achieve. Thus, this paper proposes a digital twin-enabled two-stage synchronization mechanism for upper-level muti-objective offline planning and lower-level Out-of-Order (OoO) online scheduling for logistics-assembly operations management. A numerical study is implemented for the performance evaluation of the proposed model. The experiment result demonstrates that this two-stage synchronization model can provide near-optimized solutions under stochastic environments and achieve higher resilience to reduce the fluctuations caused by uncertainties.