(cont.) Both estimated and input sell-up values are tested on AL1 whereas only input sell-up values are tested on AL2. The findings of the simulations indicate that using FP typically results in the highest revenues for AL1 among AL1 3 sell-up estimation methods. When compared against simple RM fare class threshold methods that do not consider sell-up, using FP results in up to a 3% revenue gain for AL1. Under some fare class optimization scenarios, using FP instead of input sell-up values even results in a revenue increase of close to 1%. These findings suggest that FP is robust enough under a range of fare class optimizers to be used by airlines as a sell-up estimator in unrestricted fare environments so as to raise revenues.
The book provides a comprehensive overview of current practices and future directions in airline revenue management. It explains state-of-the-art revenue management approaches and outlines how these will be augmented and enhanced through modern data science and machine learning methods in the future. Several practical examples and applications will make the reader familiar with the relevance of the corresponding ideas and concepts for an airline commercial organization. The book is ideal for both students in the field of airline and tourism management as well as for practitioners and industry experts seeking to refresh their knowledge about current and future revenue management approaches, as well as to get an introductory understanding of data science and machine learning methods. Each chapter closes with a checkpoint, allowing the reader to deepen the understanding of the contents covered.This textbook has been recommended and developed for university courses in Germany, Austria and Switzerland.
The emergence of less restricted fare structures in the airline industry reduced the capability of airlines to segment demand through restrictions such as Saturday night minimum stay, advance purchase, non-refundability, and cancellation fees. As a result, new forecasting techniques such as Hybrid Forecasting and optimization methods such as Fare Adjustment were developed to account for passenger willingness-to- pay. This thesis explores statistical methods for estimating sell-up, or the likelihood of a passenger to purchase a higher fare class than they originally intended, based solely on historical booking data available in revenue management databases. Due to the inherent sparseness of sell-up data over the booking period, sell-up estimation is often difficult to perform on a per-market basis. On the other hand, estimating sell-up over an entire airline network creates estimates that are too broad and over-generalized. We apply the K-Means clustering algorithm to cluster markets with similar sell-up estimates in an attempt to address this problem, creating a middle ground between system-wide and per-market sell-up estimation. This thesis also formally introduces a new regression-based forecasting method known as Rational Choice. Rational Choice Forecasting creates passenger type categories based on potential willingness-to-pay levels and the lowest open fare class. Using this information, sell-up is accounted for within the passenger type categories, making Rational Choice Forecasting less complex than Hybrid Forecasting. This thesis uses the Passenger Origin-Destination Simulator to analyze the impact of these forecasting and sell-up methods in a controlled, competitive airline environment. The simulation results indicate that determining an appropriate level of market sell-up aggregation through clustering both increases revenue and generates sell-up estimates with a sufficient number of observations. In addition, the findings show that Hybrid Forecasting creates aggressive forecasts that result in more low fare class closures, leaving room for not only sell-up, but for recapture and spill-in passengers in higher fare classes. On the contrary, Rational Choice Forecasting, while simpler than Hybrid Forecasting with sell-up estimation, consistently generates lower revenues than Hybrid Forecasting (but still better than standard pick-up forecasting). To gain a better understanding of why different markets are grouped into different clusters, this thesis uses regression analysis to determine the relationship between a market's characteristics and its estimated sell-up rate. These results indicate that several market factors, in addition to the actual historical bookings, may predict to some degree passenger willingness-to-pay within a market. Consequently, this research illustrates the importance of passenger willingness-to-pay estimation and its relationship to forecasting in airline revenue management.
The decision to buy a fare that is higher than the desired fare, under the situation when the desired fare is not available, is known as sell-up. Passengers' willingness to sellup can have considerable impact on airline revenues. The extent of this impact is dependent upon the method used to control booking limits and other parameters associated with passenger demand and fare structure. In this thesis we demonstrate the importance of incorporating sell-up in airline revenue management. The improvement in revenue, under various scenarios, and under various seat inventory control algorithms, is discussed. We also analyze the influence of demand factor, spill, sell-up rate and fare ratio on these improvements. A modification of the EMSRb heuristic is proposed to capture the revenue potential associated with passenger sell-up. The proposed rule increases the protection levels, obtained from the EMSRb algorithm, as long as the expected gain, from every additional seat protected, is greater than the expected loss. Unlike the existing models, the proposed rule uses expected spill to determine the expected number of passengers that would sell-up at a given demand level and sell-up rate, and then adjusts the protection levels accordingly. This makes it robust to variations in demand levels. We have also developed a simulation to compare the performance of the existing rules with that of the proposed heuristic. The simulation has the ability to account for errors in sell-up estimation and variability in demands. It is shown that the revenue gains under the proposed rule may not exist under all situations. In the tests performed in this thesis, the improvements over the original EMSRb algorithm vary from 0% to over 2.5%. Although the gains are not consistent, the proposed rule does not cause any negative impact on overall revenues and hence is unlikely to pose any risk when implemented over the original EMSRb algorithm.
Accurate forecasts are crucial to a revenue management system. Poor estimates of demand lead to inadequate inventory controls and sub-optimal revenue performance. Forecasting for airline revenue management systems is inherently difficult. Competitive actions, seasonal factors, the economic environment, and constant fare changes are a few of the hurdles that must be overcome. In addition, the fact that most of the historical demand data is censored further complicates the problem. This dissertation examines the challenge of forecasting for an airline revenue management system in the presence of censored demand data. This dissertation analyzed the improvement in forecast accuracy that results from estimating demand by unconstraining the censored data. Little research has been done on unconstraining censored data for revenue management systems. Airlines tend to either ignore the problem or use very simple ad hoc methods to deal with it. A literature review explores the current methods for unconstraining censored data. Also, practices borrowed from areas outside of revenue management are adapted to this application. For example, the Expectation-Maximization (EM) and other imputation methods were investigated. These methods are evaluated and tested using simulation and actual airline data. An extension to the EM algorithm that results in a 41% improvement in forecast accuracy is presented.
Recent technological improvements have allowed airlines to implement sophisticated Revenue Management systems in order to maximize revenues. Computational capabilities make it possible to perform network-based analysis of supply and demand and therefore to increase the gains achieved with the help of "0- D control" Revenue Management algorithms. However, the more commonly used and cheaper flight leg-based algorithms have not yet been used to the best of their potential and can still benefit from better modeling of passenger behavior. Our first purpose in this thesis is therefore to evaluate the benefits of incorporating sell-up models into current leg-based airline Revenue Management algorithms. Another question we would like to try and address is whether it would be possible to improve the leg-based models to reach revenue gains comparable to those of O-D control algorithms. To try and achieve this goal, we improve the modeling in our leg-based Revenue Management algorithms by accounting for the possibility of sellup, that is the probability that a passenger will accept a more expensive ticket than originally desired if seats are not available at the lower fare. In addition, previous research has shown that there are revenue gains to be achieved through better forecasting, therefore, we also evaluate the use of better forecasting methods and quantify their revenue impact. In particular, we focus our efforts on understanding the impact of the unconstraining models on revenue gains by using various detruncation methods and comparing their effect.
In the airline industry, spill refers to passenger demand turned away from a flight because demand has exceeded capacity. The accurate estimation of spill and the lost revenue it implies is an important parameter in airline fleet assignment models, where improved estimates lead to more profitable assignments. Previous models for spill estimation did not take into account the effects of passenger choice and airline revenue management. Since revenue management systems protect seats for later-arriving higher fare passengers, revenue management controls will influence the number of spilled passengers and their value because they will restrict availability to lower fare passengers even if seats on the aircraft are available. This thesis examines the effect of various revenue management systems and fare structures on spill, and, in turn, the marginal value of incremental capacity. The Passenger Origin Destination Simulator is used to simulate realistic passenger booking scenarios and to measure the value of spilled demand. A major finding of the research is that in less restricted fare structures and with traditional revenue management systems, increasing capacity on a flight leads to buy-down which can result in negative marginal revenues and therefore revenue losses. This behavior is contrary to conventional wisdom and is not considered in existing spill models. On the other hand, marginal revenues at low capacities are greater than would be predicted by first-choice-only spill models because some passengers will sell-up to higher fares to avoid spilling out. Additionally, because of passenger recapture between flights, adding capacity to one flight can lead to revenue losses on another. Therefore, the marginal value of incremental capacity is not always positive. Negative marginal revenues and associated revenue losses with increasing capacity can at least be partially mitigated by using more advanced revenue management forecasting and optimization algorithms which take into account passenger willingness to pay. The thesis also develops a heuristic analytical method for estimating spill costs which takes into account the effects of passenger sell-up, where previous models tend to underestimate the spill cost by only modeling passengers' first choices. The heuristic demonstrates improved estimates of passenger spill: in particular, in restricted fare structures and for moderate amounts of spill, the model exhibits approximate relative errors on the order of 5%, a factor of two improvement over previous models.
Even as the distribution and sale of commercial airline tickets has shifted in recent years from physical reservation offices to the Internet, many airline commercial processes remain highly reliant on pre-Internet technologies and standards. This legacy infrastructure compels airlines to publish a discrete set of prices in each market they serve, and to select prices for each itinerary from among only this limited set of possible price points. Recent advancements in distribution technology, such as the New Distribution Capability (NDC), offer airlines the chance to break away from these constraints. These new standards enable the creation of customized offers with prices that could be generated dynamically in real time. While airlines have shown interest in these new technologies, practical methods for integrating dynamic pricing into existing airline revenue management (RM) and distribution systems have yet to be defined and evaluated by academics or practitioners. In this work, we propose the first mechanisms for dynamic pricing designed specifically for use in the airline industry. By selectively providing increments or discounts based on demand segmentation and estimates of willingness-to-pay (WTP), our mechanisms can increase airline revenues by stimulating new bookings from price-sensitive travelers while encouraging more price-inelastic travelers to buy up to higher price points. Moreover, the methods are compatible with the pricing, RM, and distribution systems currently used by airlines today. Our dynamic pricing heuristics emerge from the development of a novel theoretical model of customer choice. Using the model, we introduce a new concept called "conditional WTP" to describe how a customer's willingness-to-pay for an itinerary can change depending on the other alternatives available in his choice set. We show how assuming an unchanging maximum WTP for air travel, as in past work on dynamic pricing, can lead to overestimation of WTP in competitive environments, and describe how an airline's estimates of conditional WTP play an integral role in our dynamic pricing mechanisms. We test our dynamic pricing methods in the Passenger Origin-Destination Simulator (PODS): a robust agent-based booking simulation that models the interactions between passengers and airlines. In a complex, competitive network, we find that our heuristics can increase airline revenues by up to 1 - 4% from traditional pricing and RM alone. Incrementing prices can result in revenue gains through an increase in yield, and discounting can lead to higher revenues through demand stimulation and share shift from other airlines. In both cases, we identify a phenomenon we call "forecast spiral-up" which increases yield by protecting more seats for higher-value fare classes. We also develop a variant of the heuristic in which multiple substitutable flights are priced simultaneously, leading to additional revenue gains. Finally, we provide the first in-depth assessment of the practical implications of dynamic pricing for the airline industry. We focus on airline concerns that dynamic pricing could lead to price wars, excessive discounting, and a race to the bottom. We also evaluate some of the potential legal implications and customer reactions that could emerge as dynamic pricing becomes more commonplace. These analyses provide new insight on how airline competition could potentially change as dynamic pricing is integrated into traditional airline processes.
Pricing is about deciding your market position whereas revenue management is the strategic and tactical decisions firms take in order to optimize revenues and profits. This book offers insights into research, theories, applications and innovations and how to makes these work in different industries.
(Cont.) The use of DPL achieves as much as 7.3% revenue improvement over EMSRb with Q-Forecasting at high demand. In contrast, the performance of DP-GVR is weaker especially against an advanced RM method, regardless of sell-up input or estimator used. On the other hand, results from a bigger network illustrate that an airline that practices DP-GVR performs much better against both simple and advanced competing RM methods. We conclude that the performance of the theoretically appealing DPL and DP-GVR depends on the environment in which they are used, the types of passenger sell-up estimator employed, as well as the Revenue Management method applied by the competitor.
