Conference Agenda

We present a data-driven approach to RM which eliminates the need for demand forecasting and optimization techniques. We develop a methodology to generate bid prices using historical booking data only. Our approach is an ex-post greedy heuristic to estimate proxies for marginal opportunity costs as a function of remaining capacity and time-to-departure solely based on historical booking data. We utilize a neural network algorithm to project bid price estimations into the future. We conduct an extensive simulation study where we measure our methodology’s performance compared to that of an optimally generated bid price using dynamic programming (DP) and compare results in terms of both revenue and load factor. We also extend our simulations to measure performance of both data-driven and DP generated bid prices under the presence of demand misspecification. Our results show that our data-driven methodology stays near a theoretical optimum (< 1% revenue gap) for a wide-range of settings.

We study a class of stochastic hidden nonconvex optimization arising from revenue management. Leveraging the implicit convex reformulation (i.e., hidden convexity), we develop stochastic gradient-based algorithms and establish their sample and gradient complexities for achieving global optimal solutions for the nonconvex problem. Interestingly, our proposed Mirror Stochastic Gradient (MSG) method operates only in the original space using gradient estimators of the nonconvex objective and achieves sample complexities, that matches the lower bounds for solving stochastic convex optimization problems. In air-cargo network revenue management, extensive numerical experiments demonstrate the superior performance of our proposed MSG algorithm for booking limit control with higher revenue and lower computation cost than state-of-the-art bid-price-based control policies.