Passenger Origin Destination Simulator Research Articles

Airline revenue management with segmented continuous pricing: methods and competitive effects

With the introduction of IATA’s New Distribution Capability (NDC), airlines will no longer be limited to discrete fare classes for their fare product distribution but could show fare quotes from continuous ranges to booking requests. NDC will also allow airlines to present different fare quotes to passengers from different demand segments as identified by the airline. In theory, airlines can better extract passenger willingness to pay, and thus, see gains in revenue, by offering segmented continuous fare quotes to different passengers requesting to book. This paper describes the revenue management (RM) methods for Segmented Continuous Pricing and examines their potential effects on airlines’ revenue through simulations in the Passenger Origin–Destination Simulator (PODS). We describe a class-based algorithm for continuous pricing, a straightforward extension from the traditional methods used with existing RM systems. Our simulation results show that in a calibrated scenario in which only one airline adopts Segmented Continuous Pricing and has an 80% accuracy in identifying business versus leisure passenger booking requests, the first-mover airline can see as much as a 17% revenue gain, at the expense of competitors. The revenue gains come primarily from the leisure passenger segment by offering lower fares than competitors closer to departure. The first-mover airline loses bookings but does not see losses in revenue from the business passenger segment. We also explore potential response strategies by the competing airlines. We discover that competitors can reverse the first-mover’s revenue gain by removing their fare restrictions while still using traditional RM methods. We conclude that although adopting Segmented Continuous Pricing is promising in theory, its gains in practice will depend heavily on the competitive situation and the responses made by competing airlines.

Dynamic offer generation in airline revenue management

The New Distribution Capability and new retailing platforms will enable airlines to respond to shopping requests with bundled offers of flights and ancillary services, representing an evolution from traditional, flight-focused optimization. Assembling an attractive set of offers to display to customers therefore represents a new joint pricing and assortment optimization problem in airline revenue management. In this paper, we introduce an initial optimization approach for the selection and pricing of a la carte and bundled flight and ancillary offers. First, we propose a customer choice model that captures the impact of ancillary bundles on flight itinerary choice. We then calculate prices for each offer from a continuous range of price points and display the offer set that maximizes expected revenue for a given customer segment. We illustrate the approach using a single-flight, single-ancillary base case and discuss extensions to more complex environments. Tests in the Passenger Origin–Destination Simulator (PODS) show that dynamic offer generation (DOG) can increase net revenue when used by one or more airlines in a competitive network, assuming that the airlines are able to accurately segment incoming requests and estimate the average willingness-to-pay of each segment. We find that the majority of the revenue gains of DOG are due to competitive effects from the dynamic pricing of the flight component of the offer. The bundling mechanism of DOG is a secondary source of revenue gain that can be realized when customers take bundled ancillary services into account when choosing the flight. Our results provide insight for practitioners that are implementing offer optimization systems and processes. For example, in line with the previous literature on bundle pricing, we find that in transparent distribution channels an ancillary service should be bundled with the flight when the valuation for the ancillary is high or when its marginal cost of provision is low. We close by discussing the strategic and managerial implications of a move from traditional distribution strategies to a next-generation, offer-focused approach.

Continuous pricing algorithms for airline RM: revenue gains and competitive impacts

Traditionally, airlines have been limited to a set of fixed fare classes and, in turn, price points, to distribute their fare products. The advent of IATA’s new distribution capability (NDC) will soon enable airlines to quote any fare from a continuous range. In theory, such continuous pricing could increase revenues by extracting more of the consumer surplus, through its ability to offer more granular fares, closer to the customer’s willingness-to-pay (WTP). In this article, we describe several algorithms that lead to the quotation of a single fare from a continuous range. These algorithms either rely on traditional fare classes for the purpose of forecasting and optimization (class-based), or completely abandon the notion of fare classes, instead assuming different WTP distributions within each booking period prior to departure (classless). We describe how these algorithms build upon and differ from their traditional RM counterparts. Performance of these heuristics is then benchmarked against traditional class-based RM, and competitive impacts are analyzed when continuous pricing is adopted by one airline asymmetrically or both airlines symmetrically in a hypothetical 2-carrier network in the passenger origin–destination simulator (PODS). We find that continuous pricing is generally revenue-positive, and the revenue gains can be as high as 2.0% for the first-mover and reach up to 1.2% when both airlines adopt the new method. In addition, we show that these gains depend on the number of fare classes in the traditional fare structure used as a baseline, and that they are smaller under lower demand-to-capacity ratios.

Can demand forecast accuracy be linked to airline revenue?

Since accurate demand forecasts are a key input to any airline revenue management system, it is reasonable to assume that an improvement in demand forecast accuracy would lead to increased revenues. However, this relationship has often been called into question. Past work has not conclusively proven that more accurate demand forecasts lead to higher revenue, causing researchers and practitioners to debate whether the concept of demand forecast accuracy itself is “myth or reality.” In this paper, we demonstrate that it is possible to consistently link demand forecast accuracy to airline revenue. After discussing why traditional demand forecast error metrics have struggled to demonstrate this relationship, we evaluate a novel conditional demand forecast error metric which compares demand forecasts to historical bookings conditional on the set of fare classes that were open at the time of booking. We prove under some mild assumptions that minimizing conditional demand forecast error will maximize revenue under any fare structure and customer choice behavior. These theoretical findings are supported by simulations in both a simple, single-leg model and in a complex multiple-airline network in the Passenger Origin–Destination Simulator. We find that price elasticity parameter bias of ± 10% can reduce revenues by up to about 1%, while price elasticity parameter bias of ± 20% can reduce revenues by up to 4%. We close by discussing the implications of the findings for revenue management practitioners.

A dynamic pricing engine for multiple substitutable flights

As enhancements in airline IT begin to expand pricing and revenue management (RM) capabilities, airlines are starting to develop dynamic pricing engines (DPEs) to dynamically adjust the fares that would normally be offered by existing pricing and RM systems. In past work, simulations have found that DPEs can lead to revenue gains for airlines over traditional pricing and RM. However, these algorithms typically price each itinerary independently without directly considering the attributes and availability of other alternatives. In this paper, we introduce a dynamic pricing engine that simultaneously prices multiple substitutable itineraries that depart at different times. Using a Hotelling line (also called a locational choice model) to represent customer tradeoffs between departure times and price, the DPE dynamically suggests increments or decrements to the prices of pre-determined fare products as a function of booking request characteristics, departure time preferences, and the airline’s estimates of customer willingness-to-pay. Simulations in the Passenger Origin–Destination Simulator (PODS) show that simultaneous dynamic pricing can result in revenue gains of between 5 and 7% over traditional RM when used in a simple network with one airline and two flights. The heuristic produces revenue gains by stimulating new bookings, encouraging business passenger buy-up, and leading to spiral-up of forecast demand. However, simultaneous dynamic pricing produces marginal gains of less than 1% over a DPE that prices each itinerary independently. Given the complexity of specifying and implementing a simultaneous pricing model in practice, practitioners may prefer to use a flight-by-flight approach when developing DPEs.

Performance of dynamic user influence strategies in PODS under seasonality and system volatility

This paper summarizes results from Passenger Origin–Destination Simulator (PODS) research on how dynamic user influence performs under seasonality and system volatility. We explore the revenue results for two different dynamic user influence (UI) strategies—unbiased and biased high–under varying demand levels, with and without hybrid forecasting, under either strong-positive or weak-positive correlation of demand. In short, dynamic UI seeks to emulate revenue management (RM) analysts’ attempts to positively influence the RM system. In this study, we use a large “international” network with 572 O-D markets and four different airlines competing for passengers, including a low-cost carrier. The revenue results show that out of 224 individual experiments, we only found eight cases (less than 4%) where there was not a revenue improvement over the comparable base case without dynamic UI. This is a ringing endorsement of the use of dynamic UI, independent of demand level, or forecasting strategy.

Simulation shows not to panic about bid price dysfunction close to departure

We have seen many times over the years where bid prices drop precipitously in the last few days of the booking process, thus potentially opening up cheap seats at a time when passengers with the greatest willingness to pay are shopping. The research question studied here is whether an airline can effectively apply a bid price floor or freeze, in the last week or two of the booking process to increase revenue in a competitive environment with four competing airlines. This paper will provide the answer to the above question in the Passenger Origin–Destination Simulator (PODS) Network U10 (a large international network with 572 origin–destination markets, including a low-cost carrier). The results show how challenging it is to find a combination of O–D optimization method, overbooking level, and time-frame that are effective. In fact, only unbucketed dynamic programming (vector) and overbooking with a floor or freeze in the last 7 days is consistently revenue positive.

Customized dynamic pricing of airline fare products

Firms practice dynamic pricing when they charge different customers different prices for the same products, as a function of an observable state of nature. In the airline industry, dynamic pricing has historically been limited by the capabilities of distribution systems, which require filing a finite set of fare products with fixed prices. However, advancements in airline distribution technology will soon allow the generation of “customized offers,” which could include a dynamically generated price. In this paper, we propose a heuristic for customized dynamic pricing of airfares when the observable state of nature includes an observation of passenger characteristics. We first introduce a general model for decoupling the customized offer generation problem from the well-studied airline revenue management problem. After generating a baseline assortment of fare products and observing the characteristics of a passenger’s request, an airline can choose to customize that passenger’s offer by dynamically incrementing or discounting the prices that would ordinarily be offered. The methodology could be applied with existing airline revenue management methods, and would be compatible with IATA’s New Distribution Capability (NDC). For implementation, we propose a straightforward heuristic approach based on simple estimates of passenger willingness-to-pay distributions. The heuristics are simulated in the Passenger Origin–Destination Simulator, a complex airline revenue management simulation environment that takes into account passenger choice and competition. The results show that the heuristics can lead to revenue gains of up to 3–4% when practiced by one airline. Furthermore, the heuristics remain revenue positive in competitive environments.

Incorporating ancillary services in airline passenger choice models

This paper develops a new integrated passenger choice model to explain how consumers select ancillary services in conjunction with an airline itinerary and specific fare class. We propose that consumers are either classically rational, which we term simultaneous, or boundedly rational, which we term sequential. Simultaneous consumers are modeled to choose an itinerary, a fare class, and a set of ancillary services at the same time; sequential consumers are modeled to choose an itinerary and a fare class in one phase and then choose a set of ancillary services in a second phase. The sensitivity of revenues and bookings to various ancillary fee structures with the integrated choice model are illustrated by simulations in the Passenger Origin–Destination Simulator, showing that ancillary bundling can result in buy-up or buy-down by simultaneous passengers, but that sequential passengers provide more revenue to airlines.

Do hybrid forecasting and forecast multipliers still work in a “perfectly tuned” PODS international network with four competing airlines?

We have seen many results over the years where various ‘intelligent aggressiveness’ levers (e.g., forecast multipliers, aggressive unconstrainers, hybrid forecasting, and/or fare adjustment) have been used effectively by an individual airline in a competitive environment to increase its revenues. The research question studied here is whether an airline can still use any of these methods to increase revenue in a competitive environment where all the competing airlines are “perfectly” aggressive. This paper will answer the question in the Passenger Origin–Destination Simulator Network U10 (a large international network with 572 origin–destination markets with four airlines competing for passengers, including a low-cost carrier).

Tests of revenue management performance under different demand correlation assumptions

This article summarizes results from Passenger Origin–Destination Simulator (PODS) research on how a revenue management (RM) system performs under various assumptions about demand correlation (between early- and late-arriving customers). PODS typically assumes a relatively strong positive correlation; this article shows whether and how much the results change under weak positive correlation. First, we explore the revenue results for three different dynamic user influence (UI) strategies – unbiased, biased low and biased high – under varying demand levels, with and without hybrid forecasting (HF). In short, dynamic UI seeks to emulate RM analysts’ attempts to positively influence the RM system and was originally presented by Hao at a 2013 PODS meeting. Next, we explore the revenue results for three different Origin–Destination (O–D) strategies under varying demand levels, with and without HF for different correlated demand scenarios. In both studies, we use a large ‘international’ network with 572 O–D markets with four airlines competing for passengers, including a low-cost carrier.

Demand driven dispatch and revenue management in a competitive network environment

Demand Driven Dispatch (D3) is the reassignment of aircraft to flights close to departure to improve operating profitability, primarily by utilizing improved knowledge of expected demand from the airline’s revenue management (RM) system. Previous studies of D3 have not incorporated competition and have typically ignored or significantly simplified RM. In this study, the implementation of D3 is tested with complete representations of RM systems in a network environment with competition. Results are from the Passenger Origin Destination Simulator, where stochastic demand by market chooses between competing airlines with alternative schedules and fare products. Findings include important competitive feedback effects from D3 and insights about relationships between D3 and both RM and pricing. Our findings indicate that the benefits of D3 can be estimated at operating profit gains of 0.04–2.03 per cent, revenue gains of 0.02–0.88 per cent, and changes in operating costs of −0.08 to 0.13 per cent. However, use of D3 may harm competitor airlines more than it aids the implementer. D3 swaps early in the booking process can lead to heavy dilution. Late swaps lead to smaller increases in loads but substantial increases in revenue. The relationship between revenue-maximization and cost-minimization in profit-maximizing D3 is highly influenced by the timing of swaps and revenue estimation.

Dynamic pricing – The next revolution in RM?

Existing revenue management systems (RMS) base their recommendations on historic observations and do not explicitly consider competition. This means that RMS recommendations often are not appropriate for real-time competitive situations. Dynamic pricing (DP) is an extension of RMS that dynamically calculates the optimal price, taking into account the airline’s strategy, customer-specific information and real-time alternative offerings. By optimizing the contribution within the shopping session, DP has a more current and detailed view of demand and can improve RMS performance. We investigated the performance of DP using two simulators, Altea Benchmarking Engine and Passenger Origin Destination Simulator and demonstrate that DP can deliver substantial revenue benefits with no modification to existing revenue management (RM) processes. However, the deployment of DP into the airline distribution process will be a challenge, because it affects all shopping and downstream processes, such as ticketing, servicing, revenue accounting, RM and interline settlement, that rely on information from existing fare aggregators. Nevertheless, the potential benefits of DP are so compelling that we believe the effort to bring this technology into practice is warranted.

Personalization in airline revenue management – Heuristics for real-time adjustment of availability and fares

Through improvements in information technology and distribution, airlines are increasingly gaining access to detailed information about potential passengers at the time of booking. Airlines could use this passenger information to provide availability or a fare quote that is tailored specifically toward the passenger that is making the request. For instance, IATA’s New Distribution Capability (NDC) will allow airlines to offer passengers a personalized fare offer that could include a customized price. By strategically targeting personalized offers toward the right customers, airlines could gain new bookings from price-sensitive passengers while encouraging more price-inelastic travelers to buy-up to higher price points. This paper, the second in a series on dynamic availability in airline revenue management, proposes two heuristics that could be used to provide targeted availability or fare offers to specific customers or segments of customers after making a simple estimate of the passenger’s willingness-to-pay (WTP). Tests in the Passenger Origin–Destination Simulator (PODS) show that the heuristics are generally revenue-positive for airlines and are stable in a competitive revenue management environment. Of particular interest is “dynamic discounting,” which increases both yields and load factors by offering targeted discounts to leisure passengers in specific situations. The results improve on past work on dynamic availability and introduce new possibilities for personalization in airline revenue management.

Intelligent Aggressiveness: Using Forecast Multipliers, Hybrid Forecasting, Fare Adjustment, and Unconstraining Methods to Increase Revenue*

ABSTRACTMany studies have begun the exploration of airlines using intelligent aggressiveness (IA) in unidimensional directions (e.g., forecast multipliers alone). This article uses the sophisticated passenger origin–destination simulator (PODS) to examine the revenue impact of four different IA levers—forecast multipliers, unconstraining, hybrid forecasting (HF) and fare adjustment (FA). We also explore the impacts in two different origin–destination networks. Due to the competitive nature of PODS (two or four airlines competing) and its allowance for customer choice, we are able to assess all the implications, including the impact of spill, upgrades and recapture. We find that with a single IA lever, independent of the network and demand level, in a more‐restricted fare environment, the optimal lever is almost always HF with moderate‐to‐aggressive estimates of willingness‐to‐pay, with revenue gains of 0.4–4.3% in a large global network, and gains of 1.7–4.2% in a domestic network, depending on demand level and optimization method used. We also test two additional, less‐restricted fare environments and find that revenue improvements have a wider range (0.8–6.3%) with a single lever in the larger network. Finally, we explore the impacts of allowing the competitors to use basic IA and the airline of interest to use multiple IA levers.

Dynamic availability of fare products with knowledge of customer characteristics

Recent advancements in airline distribution technology, including IATA’s New Distribution Capability (NDC), could offer airlines a wealth of information about the characteristics of individual customers arriving to book. Recent literature has envisioned a world where airlines, enabled by this new distribution technology, are able to make personalized fare quotes in real time, effectively heralding the end of traditional booking classes. However, academic literature suggests that this type of price discrimination may intensify competition and lead to lower revenues for firms. In this article – the first in a series of papers exploring models of dynamic product availability with knowledge of customer characteristics – we test a simple dynamic availability (DynA) heuristic that adjusts fare class availability after inheriting information about a customer’s trip purpose. Through results from the Passenger Origin-Destination Simulator, we find that this DynA heuristic is often revenue-negative for airlines and unstable in competitive environments. The results raise questions about airline benefits from simple approaches to personalized pricing enabled by NDC, and open up opportunities for future research regarding more nuanced and advanced strategies for customized pricing and availability in the airline industry.

Bid-price behavior in network RM: Forecast decrementing and bid-price bounding

This article investigates two practical concerns with bid-price control origin–destination (OD) revenue management (RM) systems. First, we examine forecast decrementing methods for the common situation where demand forecast generation is decoupled from booking policy optimization. Second, we explore the impacts of two bounding schemes designed to eliminate bid price declines close to departure. Using the Passenger Origin–Destination Simulator (PODS), we find (i) more sophisticated forecast decrements increase revenue compared with the traditional approach when combined with advanced forecasting methods incorporating passenger willingness to pay and (ii) bid price bounds increase revenue with standard forecasting methods but decrease revenue with advanced forecasting methods.

Optimization models in RM systems: Optimality versus revenue gains

Optimization models in airline revenue management (RM) systems have evolved from single flight leg to network revenue maximization to marginal revenue optimization for less restricted fare structures. This article reviews the most common optimization approaches that have been widely implemented in airline RM systems, with a focus on how the mismatch between model assumptions and reality can affect achievable revenue performance. Simulation findings from the Passenger Origin-Destination Simulator are used to illustrate how robustness and revenue gains, as opposed to theoretical optimality, have driven the widespread adoption of practical optimization models in RM systems.

Implementation of MNL choice models in the passenger origin-destination simulator (PODS)

In this article, we describe the implementation of a choice-based forecasting methodology in the Passenger Origin-Destination Simulator. The parameters of a MNL choice model (fare and restrictions) are estimated for each of the 482 markets of a large domestic network at the various stages of the booking window. These estimates are then used to produce a new MNL demand forecast at the path level. Revenue results are presented when comparing this new MNL forecaster versus existing methods like leg standard, path standard and hybrid forecasting, under both leg and network optimization.

Intelligent aggressiveness: Combining forecast multipliers with various unconstraining methods to increase revenue in a global network with four airlines

Airlines are always searching for ways to maximize revenues in the hyper-competitive environment that they exist in. This article uses the sophisticated Passenger Origin-Destination Simulator (PODS) simulator to examine the revenue impact of four levels of forecast multipliers (FM) in combination with three different methods of unconstraining – Expectation Maximization (EM), Projection Detruncation (PD) and Booking Curve (BC). Owing to the competitive nature of PODS (four airlines competing for customers) and its allowance for customer choice, we are able to assess all the implications of these FM levels in combination with unconstraining, including the impact of spill, upgrades and recapture. We find that in this fully/semi-restricted fare environment, under leg optimization and either EM or PD, the optimal level is generally FM of 1.1, independent of demand level, and that under BC unconstraining, the optimal level is FM equal to 1.2. When using realistic booking data from major global airlines to calibrate PODS’ largest global network (U1), we show that becoming more intelligently aggressive with FM can lead to revenue gains of 0.2–0.6 per cent under EM or PD unconstraining and gains of 0.3–1.2 per cent under BC unconstraining. Under network optimization, it is best not to use FM under any unconstraining method, independent of demand level.