Pricing-based revenue management for flexible products on a network

This paper proposes and analyses a pricing-based revenue management model that allows flexible products on a network, with a non-trivial extension to group reservations. Under stochastic demand the problem can be solved using dynamic programming, though it suffers from the curse of dimensionality. The solution under deterministic demand gives an upper bound on the stochastic problem, and serves as a basis for two heuristics, which are asymptotically optimal. Numerical studies, based on a problem instance from practice, show that the heuristics perform well, even under uncertainty in demand. Moreover, neglecting flexible products can lead to substantial revenue loss.