Dynamic pricing and inventory control with learning (Q4533476)

The authors determine optimal pricing and stocking policies over time for a monopolist when a given market parameter is initially unknown, and finally revealed through learning. They find that the first-period optimal price increases with the length of the planning horizon. However, for a given problem horizon, prices can rise or fall over time, depending on how the scale parameter influences demand. Moreover, they provide the characterization of the optimal stocking quantity decision, and a computational algorithm to obtain the optimal solution. As stated in the concluding comments, in reality, the ``randomness is incorporated in \(\varepsilon\) such that the probabilistic nature of the problem does not end at the first occurrence of leftovers.'' In addition, they present a dynamic model that simultaneously links price, learning from sales, and inventory. However, there is no inventory cost involved. Consequently, I conclude that it is an interesting and relevant basic research paper. However, it needs more further study to make it applicable in the real world.