摘要：We consider a fundamental problem, called \emph{QoS-aware Multicommodity Flow}, for assessing robustness in transportation planning. It constitutes a natural generalization of the weighted multicommodity flow problem, where the demands and commodity values are elastic to the Quality-of-Service (QoS) characteristics of the underlying network. The problem is also fundamental in other domains beyond transportation planning. In this work, we provide an extensive experimental study of two FPTAS for the QoS-aware Multicommodity Flow Problem enhanced with several heuristics, and show the superiority of a new heuristic we introduce here.
