Surveys of fragmentations, especially in the early stages of the given event, are fundamental for determining the number of fragments, identifying and cataloging them, and monitoring their future evolution. The development of a ground-based optical survey strategy, i.e., a suitable observation and detection method for the fragments generated by these events, is an important contribution to acquiring data and monitoring these catastrophic phenomena. An optical survey offers an interesting and cost-effective method that supports radar operations in the Low Earth Orbit regime and can monitor higher orbits where radar cannot be used. This paper presents a developed optical survey strategy for multi-observatory observations. The strategy was tested on the fragmentation event of FREGAT R/B CLUSTER 2, a rocket body with a “dummy” payload, fragmented on 8 April 2024 on a Highly Elliptical Orbit. The observational campaign involved different observatory systems, and it represented a key collaboration within the Inter-Agency Space Debris Coordination Committee. The survey started from a simulation of the cloud of fragments and was implemented by the planification and coordination of different observatory systems with different schemes and methods to scan the sky vault. The acquired survey data were analyzed using machine learning methods to identify the unknown objects, i.e., the fragments. The data acquired were compared with the simulated cloud used for the survey, and a correlation of measurements belonging to the same object was performed. Also, the parent body was characterized in its tumbling motion by the light curve acquisition.