Loading...

Dynamic Pricing of Perishable Asset Using Demand Learning

Eslami Shahrbanki, Behrouz | 2008

667 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 39505 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Hajji, Alireza
  7. Abstract:
  8. This research deals with the problem of dynamic pricing of perishable assets. In this problem there are two sources of randomness: the arrival rate of customers and their reservation prices. In most studies considering this problem, it’s assumed that process of arrivals of customers follows a Poisson process with a given intensity. Thus this process is assumed to have independent increments and the information regarding the arrival times of previous customers doesn’t have any influence on the distribution of arrival times of future customers. In some recent studies it’s assumed that customers’ arrivals follow a conditional Poisson process with an unknown intensity. The distribution of this unknown parameter is then updated according to the arrival time of each customer. However, none of these studies consider the simultaneous updating of arrival rate and reservation prices. In the current study, the distributions of both arrival rate and reservation prices of customers are updated based on the observations made by the seller during the selling season. The model proposed in this study is an extension to a model in which demand learning based on the arrival times of customers is included. In order to update the distribution of reservation prices, one of the parameters of this distribution is assumed to be unknown and treated as a random variable. After the arrival of each customer, the distribution of this variable is updated using the Bayesian learning mechanism based on the response of customer to the announced price. Since the direct use implementation of Bayesian method significantly adds to the complexity of problem, an approximation method is employed for determining the posterior distribution of reservation prices. Numerical experiments indicate that the proposed model dominates the original model in all cases. This dominance is especially significant when there isn’t an accurate initial estimate for the unknown parameter or when the seller is less confident about the accuracy of this initial estimate. Based on these experiments, increasing the learning time will lead to a significant increase in relative efficiency of learning model only if the seller has a firm belief in his inaccurate initial estimate. It’s also known from the numerical experiments that an increase in variance of distribution of unknown parameter usually results in learning with a faster pace
  9. Keywords:
  10. Revenue Management ; Yield Management ; Dynamic Pricing ; Bayesian Learning ; Distribution Function Approximation

 Digital Object List

  • محتواي پايان نامه
  •   view

 Bookmark

No TOC