Motivated by Boeing 777 fully autonomous upright build project, orchestration of human and robotic agents are studied. Tasks must be precisely allocated, sequenced, and coordinated among agents subject to temporal and spatial constraints. The problem is formulated as a flexible job shop with sequence-dependent setup to capture heterogeneous agents and travel time. Two exact central approaches are proposed: a mixed integer programming and a constraint programming, and tested for real-time perspective. The computational study demonstrates the proposed method can generate optimal schedules up to 100 agents with 1000 subtasks that requires 10 subtasks per agent on average, within 183 seconds, a substantial improvement over all other benchmark approaches in the literature.