The Earth Observation (EO) market is rapidly growing due to technology miniaturization, cheaper launch opportunities and wider spectrum of EO applications. Along with an exponential growth of the ground based users that can access Low Earth Orbit (LEO) spacecraft data, this growing community represents an important demand for Data Relay missions. LEO spacecraft have short visibility windows to the Ground Stations (GS) which limit their throughput. Data Relay missions are comprised of spacecraft at higher altitude orbits (Geostationary Orbits) acting as relays of data among LEO spacecraft and GS. Those missions are then able to offer more frequent data downlink opportunities to the LEO spacecraft thus increasing the volume of the data reaching the ground and improving the responsiveness between users' requests and downlink operations. Ground based Mission Planning Systems (MPS) are commonly managing such complex missions, representing a large operational cost. In this paper, we propose the application of a Swarm Intelligence algorithm to the design of an automated MPS for Data Relay missions. Automated MPS have the potential of saving operational costs while leaving the high level decisions to human operators. This paper represents the first time that an Ant Colony Optimization (ACO) algorithm is applied to this type of scheduling problem. This family of algorithms is generally found to offer good level of efficiency and scalability. In this work, we compare our approach against an algorithm popular in literature, called Squeaky Wheel Optimization (SWO) and show how our ACO algorithm outperforms it.