The experiments we describe here show that summary information improves performance significantly when tasks within the same hierarchy have constraints over the same resource, and solutions are found at some level of abstraction. At the same time, there are cases where abstract reasoning incurs significant overhead when solutions are only found at deeper levels. However, in domains where decomposition choices are critical, we show that this overhead is insignificant because the CFTF heuristic chooses decompositions that more quickly lead to solutions at deeper levels. These experiments also show that the EMTF heuristic outperforms a simpler heuristic depending on the decomposition rate, raising new research questions. We use the ASPEN Planning System [Chien, Rabideu, Knight, Sherwood, Engelhardt, Mutz, Estlin, Smith, Fisher, Barrett, Stebbins, Tran, 2000b] to coordinate a rover team for the problem described next.