Heterogeneity in Preferences of Air Travel Itinerary in a Low-Frequency Market

Abstract

Research in the choice behavior of air travelers has evolved to include an analytical focus on variation in the sensitivities of travelers to factors influencing itinerary choice. Some choice studies have moved beyond a focus on assumed representative, mean-level sensitivities toward a goal of representing the distribution of preferences across a sample. The mixed multinomial logit model has served as a valuable means of estimating such distributions of air travel preferences, including studies of business travelers, impacts of airport level-of-service attributes, distributions of willingness-to-pay (WTP), and information-processing strategies. Does the insight gained in previous studies, focusing on preferences in mature markets with relatively high per capita rates of air travel activity, apply to markets with low frequencies of airline patronage? This study centers on a survey of travelers in Tehran, Iran, a low-frequency air travel market. The analysis incorporates tests of a full range of distributions of random parameters to determine whether the impacts of restricting distributions allow only normality and confirms the potential to improve model fit with alternative distributions. The estimated distributions of WTP measures confirm the value of accounting for preference heterogeneity in the analysis of choice of air itinerary behavior in a low-frequency market and yield lower mean WTP values relative to analysis that omits the effects of preference heterogeneity.

Accounting for Systematic Influences on Preference Heterogeneity in Air Travel

“…Indeed, the results show that those who have more general travel experience are more sensitive to travel time and that more experienced business travelers are more sensitive to airline fare. These results are intuitively reasonable and consistent with those presented by Rezaei et al (5). However, more experienced passengers between a specific O-D pair reveal less sensitivity to the airline fare and more sensitivity to the FFP services (e.g., use of special waiting lounge services in the connecting airport, use of free taxi services) in choosing an option for traveling between that O-D pair.…”

“…The findings suggest that there are large, statistically significant differences in coefficient values of airline fare, weekly flight frequency, travel time, network, and FFP between segments when respondents are segmented by trip purpose. Rezaei et al (5) give more information about the segmented models by sociodemographic and trip-related attributes. All coefficients of flight attributes were tested to check whether the random distribution was the most appropriate representation.…”

“…In previous work (5) an attempt was made to handle systematic heterogeneity across passengers in the context of fixed preference parameters by using segmentation by sociodemographic and triprelated attributes (deterministic segmentation strategy). However, a challenge in this approach is identifying the most appropriate means of defining segments for variables of interest (6).…”

Air itinerary choice in a low-frequency market: A decision rule approach

“…Incomplete or inadequate collected data (e.g., data of passengers who chose routes rarely considered by other passengers) were removed from the sample. After these screening processes, the number of usable responses declined to 421 (Rezaei et al, 2011).…”

Analysis of Crash Severity of Texas Two Lane Rural Roads Using Solar Altitude Angle Based Lighting Condition

“…It is worth mentioning that the developed version of the ordered models, such as the generalized ordered logit/probit model, is an efficient solution to lessen the parallel assumption [26,30], although it still exists [11]. Furthermore, standard logit/probit models are limited to a specific distribution and cannot find heterogeneity in individuals behavior nor even its sources [9,31]. Due to the ordered probit model s weaknesses, some researchers have used multinomial logit models (MNL) to analyze the crash injury severities [19,22].…”