The use of formation flying of distributed space systems, especially for space-based interferometry, is receiving much attention by mission designers. Optimal maneuver planning for these missions is critical to ensure the safe operation of the spacecraft and to maximize the mission science returns. One such mission, Darwin, requires complex observation scheduling complicated by a number of interconnected temporal constraints placed on the mission. Following a defined maneuver planning architecture this paper introduces a science operations planner that helps maximize observation time through science operations schedule optimization. Comparison of this method with a simple benchmark planner is given and shows that schedule performance increases of up to 12% can be achieved. Though these increases can only be achieved using significantly more computational resources than the benchmark planner, they are found using constraints that would allow the planner to be able to operate autonomously onboard one of the formation spacecraft. Copyright © 2009 by Cranfield Univ.