作者: Feng Liu , Ying Qu , Yanan Wang
DOI: 10.1007/978-3-642-32054-5_194
关键词:
摘要: By dividing customers into different segments according to origin destination (O-D) pairs, we consider a network capacity control problem where each customer chooses the open product within segment he belongs to. Starting with Markov decision process (MDP) formulation, approximate value function an affine of state vector and develop our model based on O-D demands. The approximation results in linear program (LP). We give column generation procedure for solving dual LP provide numerical results.