Computer Science – Learning
Scientific paper
2011-01-24
Computer Science
Learning
Scientific paper
In this work, we consider a retailer selling a single product with limited on-hand inventory over a finite selling season. Customer demand arrives according to a Poisson process, the rate of which is influenced by a single action taken by the retailer (such as price adjustment, sales commission, advertisement intensity, etc.). The relation between the action and the demand rate is not known in advance. The retailer will learn the optimal action policy "on the fly" as she maximizes her total expected revenue based on observed demand reactions. Using the pricing problem as an example, we propose a dynamic "learning-while-doing" algorithm to achieve a near optimal performance. Furthermore, we prove that the convergence rate of our algorithm is almost the fastest among all possible algorithms in terms of asymptotic "regret" (the relative loss comparing to the full information optimal solution). Our result closes the performance gaps between parametric and non-parametric learning and between a post-price mechanism and a customer-bidding mechanism. Important managerial insights from this research are that the value of information on 1) the parametric form of demand function and 2) each customer's exact reservation price are rather marginal. It also suggests the firms would be better off to perform concurrent dynamic learning and doing, instead of learning-first and doing-second in practice.
Deng Shiming
Wang Zizhuo
Ye Yinyu
No associations
LandOfFree
Close the Gaps: A Learning-while-Doing Algorithm for a Class of Single-Product Revenue Management Problems does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Close the Gaps: A Learning-while-Doing Algorithm for a Class of Single-Product Revenue Management Problems, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Close the Gaps: A Learning-while-Doing Algorithm for a Class of Single-Product Revenue Management Problems will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-541141