REVENUE MANAGEMENT: GENETIC OVERBOOKING/ALLOCATION  ALGORITHMS FOR THE HOTEL INDUSTRY

Victor Pimentel

The extant literature on hotel revenue management has optimized overbooking limits separate from optimizing allocation limits. That is, the number of booking to accept in excess of the hotel’s physical capacity on a given night is determined independently from how to allocate that reservation space between customer groups with different willingness-to-pays (i.e., rate classes). This paper develops a new optimization approach that does overbooking and allocation simultaneously. Chapter Two via a simulation model of a hotel reservation system that has all of the real world dimensions to it, we determine that a worst-case net revenue (room revenues net of overbooking penalties) improvement of 10.7% can be obtained by switching to our simultaneous method. We also determine that our method’s relative improvement advantage is greatest when the demand-to-capacity ratio (i.e., demand intensity) is lower and the gap in timing between reservation requests and room rates between the rate classes is higher. Finally, we demonstrate that our new approach is computationally feasible in practice since is can be executed in less than two minutes on a realistic-sized problem. Chapter Three uses explicit price optimization where overbooking and allocation are done simultaneously. For single room at a time reservations only. This has never been done before in an explicit price optimization setting. Thus, we will quantify the revenue improvement of doing overbooking and allocation jointly in this environment. Chapter Four focus on prioritizing how a total furcating system should be parameterized: how much data to use to update each specific parameter, and how long to freeze each parameter/forecast before updating. This paper fills this prioritization void by utilizing a full-functionality hotel reservation system simulation as the basis for running screening experiments on an exhaustive set of forecaster parameters on their impact on bottom-line system performance (average nightly net revenue). A screening experiment is run for each general type of operating environment (demand intensity, degree of market segment differentiation) that a property might face.

REVENUE MANAGEMENT: GENETIC OVERBOOKING/ALLOCATION ALGORITHMS FOR THE HOTEL INDUSTRY

Files and links (1)

Abstract

Metrics

Details