Parametric demand learning with limited price explorations in a backlog stochastic inventory system

Phase Transitions and Cyclic Phenomena in Bandits with Switching Constraints

“…Dynamic pricing with demand learning has proven its effectiveness in online retailing. However, it is well known that in practice, sellers often face business constraints that prevent them from conducting extensive price experimentation and making frequent price changes, see Cheung et al (2017) and Chen and Chao (2019). The seller's sequential decision-making problem can be modeled as a BwSC problem, where changing from each price to another price incurs some cost, and there is a limit on the total cost incurred by price changes.…”

A heuristic policy for dynamic pricing and demand learning with limited price changes and censored demand

“…The studies that are most relevant for the work presented in this paper are [1], [9], [10], [11]. In [1] the authors study a dynamic pricing problem with limited price changes but there is no demand censoring and no inventory decisions.…”

“…The authors present a policy for this problem and characterize the regret as a function of the number of price changes. In [11], a multi-period stochastic inventory system with backlogs and demand uncertainty is considered. Although [11] do consider a setting with limited price changes, they make the assumption that potential demand (and thus the lost sales) is observed and can be backlogged.…”

“…• The seller decides to increase the inventory level to y t ∈ Y and decides on the price p t that will be charged in period t. Similar to previous studies (e.g. [12], [13], [11]), we assume that replenishment occurs instantly without any delay. • The demand during period t, denoted by d t (p t ) is realized and the seller tries to satisfy as much of this demand as possible using the inventory available during the period.…”

Network Revenue Management with Limited Switches: Known and Unknown Demand Distributions

“…However, not all retailers have the infrastructure to query the realized demand in real-time, to adjust its decisions instantaneously, or to switch between actions as freely as possible, because changing the posted prices is too costly for many retailers (Levy et al 1998, Zbaracki et al 2004, and frequent price changes may confuse the customers (Jørgensen et al 2003). A common practice for many firms is that they restrict the number of price changes to be as few as possible; see Netessine (2006), Chen et al (2015), Cheung et al (2017), Chen and Chao (2019), Simchi-Levi and Xu (2019), Perakis and Singhvi (2019).…”