Loading...

Developing a Decomposition Approach for Network Revenue Management Considering Reference Price: Focusing on Hotel Industry

Aslani, Shirin | 2014

514 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 46012 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Modarres Yazdi, Mohammad
  7. Abstract:
  8. This dissertation develops a new approach in network revenue management problems, with concentration on hotel industry. Several methods and heuristics are suggested in the literature to obtain an appropriate solution for network problems, however their focus is mostly on the transportation carriers. These heuristics do not seem to have satisfying effectiveness for hotel industry cases, because of significant difference between these two industries including the network structures, demand distributions and competition markets. On the other hand, with the tremendous growth of the internet and online sale channels, customers’ purchase behavior is modifying drastically. However this behavioral change is not studied in revenue management literature. The ability of customers to get detailed information about previous sales creates an expectation for customers called reference price. This new dimension in revenue management has great impacts on customers’ willingness to pay. We investigate this behavioral aspect of customers and show that the structural properties of value function is not the same as the classic models with no reference price. Similarly the effectiveness of booking limit control policy declines significantly in the presence of reference price. However, under certain assumptions including the Poisson arrival process of customers bid price policy still leads to the optimal solution, although with a different price threshold. Obviously the inefficiency of booking limit policy, decreases the effectiveness of many heuristics that are suggested in the literature for the network problems. Furthermore, this dissertation studies the structural properties of airline and hotel industries and the decision making process of network revenue management which leads to some unfair pricing decisions by the airline carriers. We also suggest a new decomposition approach by considering the customer psychology and reference price effect. This approach converts the multiple-night stay problem into separate single-night problems by defining an effective arrival rate for each night. The effective arrival rate of each night is calculated based on the mutual specification of different nights (such as the price and availability of rooms). The decomposition approach is developed with the rolling horizon structure to be able to optimize real word problems. The suggested method helps the manager to price the rooms for one year in advance, in a timely manner. Toward the completion of this dissertation, we develop a comprehensive optimizer and simulator noted “SimML” to study and illustrate the results and to compare the performance of suggested models and approaches with the previous ones
  9. Keywords:
  10. Hotel Industry ; Pricing ; Dynamic Programming ; Revenue Management ; Reference Price

 Digital Object List

 Bookmark

No TOC