Estimating Demand for a New Regional Transport Aircraft

Executive Summary

This report seeks to gauge the extent of potential demand for a new low-cost civilian air carrier serving the need for air travel between metropolitan cities, smaller cities and towns in India. More specifically, we are interested in analyzing passengers travel preferences in the aggregate market so as to conceptualize an appealing new offering that can generate primary demand. further analysis will then be done to estimate the potential primary demand generated as well as to identify attractive segments. We do so using sample data from a questionnaire administered to 487 respondents in Rajahmundry. Choice-based Conjoint Analysis is the overarching analysis method employed.  

Overall, results from analyzing our sample data finds little support for the creation of an attractive air travel offering that can generate significant primary demand. However, this might be due to an unrepresentative or biased sample. In addition, our models were noted to have low performance, measured by a low “hit ratio”. While an initial segmentation was executed on our sample subjects, we found little statistical difference between the segments along key demographic dimensions. In addition, the use of key demographic dimensions was unable to accurately predict segment members. This might hint at the lack of meaningful differentiation between our segments.

Our key recommendation is for further exploratory research to uncover other relevant variables critical to this analysis. This might help resolve the issues of (a) Model Low hit ratios where the model’s predictions often do not align with actual observed choice (b) statistically insignificant results  and (c) intuitively questionable results. Some areas to examine include further testing for a statistically significant relationship between fare and utility (specific to transport modes), which is important for pricing strategy. In addition, more effective segmentation results and/or prediction variables for segments should be found. Finally, previously eliminated variables that intuitively appears salient, such as travel duration and accessibility-convenience, can be re-tested for statistical significance.

Modelling Consumer Utility for Transport Modes - Specification 1

        In order to determine the viability of using aircraft as a mode of regional transportation, it is important to understand potential demand for this mode of transport. This can be done through Choice-Based Conjoint Analysis, in particular through using the Multinomial Logit (MNL) model to analyse the data that has been collected through surveys. Here, utility is used here as a proxy value to gauge demand amongst transport modes. In order to compute the expected utility that each mode of transport provides, the three key variables that will be examined will be the mode of transport, the fare, as well as next-day availability. Other potentially relevant factors, such as travel duration and accessibility- connectivity, were initially identified but found to be non-significant in a preliminary conjoint exercise. The details of our MNL model output is shown in Figure 1 under the appendix. Using the coefficient estimates produced, we formulate the following models for calculating utility by mode of transport :