Optimal capacity decisions of airlines under supply-demand equilibrium

Khorasani, I 2019, Optimal capacity decisions of airlines under supply-demand equilibrium, Doctor of Philosophy (PhD), Business IT and Logistics, RMIT University.


Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Khorasani.pdf Thesis application/pdf 3.93MB
Title Optimal capacity decisions of airlines under supply-demand equilibrium
Author(s) Khorasani, I
Year 2019
Abstract In the last three decades, airlines across the globe have experienced significant incidents and milestones such economic recessions, de-regulations, and jet fuel fluctuations, leading to many consolidations and even bankruptcies. Airlines seem to have a few options to respond to these disruptions and fluctuations. Capacity planning is one of the key tools that airlines apply to manage air traffic demand and their operating costs. As such, the carriers may alter the number of flights, use different types of airplanes, upgrade the seats in the aircraft, and even increase the load factor to maintain their market share and profitability, which can occasionally lead to passenger dissatisfaction. 'Capacity Planning' is defined in this research as a combination of the number of flights and aircraft size that airlines choose to manage traffic demand on a given origin-destination route. It affects the airlines' service quality and operating costs, in turn, influencing their market share and profitability. Capacity planning has become more important for airlines due to the diminishing relative significance of traditional tools such as airfare management or hedging contracts.

However, capacity planning seems to be a difficult decision-making task for airlines as they need to consider many factors on both sides of the supply-demand equilibrium of the flight market and different limitations such as access to specific aircrafts, airports, or even flight regulations. Any changes in the capacity would trigger a sophisticated set of interrelated changes in passenger demand, flight frequency, aircraft size, airfare, and flight delay, finally leading to an equilibrium shift. This statement considers economies of density that means, given no congestion, more density in terms of higher passenger demand leads to more plane-miles by either more flights or larger aircrafts. In fact, with no capacity constraints, there is an ongoing loop causing higher density from the demand side and more plane-miles from the supply side of the flight equilibrium. However, this picture is no longer valid once the capacity constraint is added to the equilibrium. Capacity constraint introduces a new player, flight delay, to the equilibrium. In other words, higher density leads to more flight delays because of capacity constraints. Flight delays bring extra costs to airlines, diminishing economies of density. Therefore, airlines need to consider all these interrelated interactions to make efficient capacity plans on their operating networks.

This thesis develops an optimisation model to assist airlines to make the optimum capacity decisions for individual routes of a given market such as a specific airport or network to maximise the potential passenger demand under the flight supply-demand equilibrium. To address this research, three key questions are identified as follows: What are the key determinants of airlines' capacity decisions under the supply-demand equilibrium of flight market? How does an airline's capacity decision influence flight delays? How can airline capacity decisions be optimised for the individual routes of a given market to maximise the total potential flight demand with respect to the market's capacity constraints? Furthermore, this research answers some significant questions related to the interactions among the key players of the supply-demand equilibrium of the flight market.

To answer these questions, this research is implemented in three steps. In the first step, the key drivers of capacity planning and demand modelling are statistically identified on both sides of the supply-demand equilibrium by applying the two-stage least square technique on the time-series cross-sectional data of 21 major routes of the Australian domestic market. In the second step, the impact of changes in the elements of capacity decisions in flight delay are investigated by using the Hausman-Taylor regression technique on the Australian domestic data. By connecting the findings of step 1 and 2, a research framework is created to be used as the basis of the optimisation algorithm in the final step. The model is developed by the inclusion of a series of exogenous and endogenous factors under the supply-demand equilibrium. To address the simultaneity among the variables, a system of four non-linear equations, flight demand, flight frequency, aircraft size, and flight delay, is developed and estimated individually by two statistical simultaneous techniques - three-stage least square technique (3SLS) and maximum likelihood estimator (MLE). The data of seven Australian domestic routes, linking Melbourne to other major cities in Australia, was applied, as the case study, to estimate the model's coefficients. Finally, the non-linear optimisation technique was applied to the estimates of 3SLS and MLE separately to find the optimum capacity plan of the given routes. All proposed models were verified and tested in different steps. As the key contribution, this thesis proposes an optimisation model based on a system of non-linear equations of the flight supply-demand equilibrium to maximise the potential flight demand of a given market with respect to the market's capacity constraints. This model is based on the theory of economies of density and applied the time-series cross-sectional data of flight market to empirically estimate the coefficients of passenger demand equation as the objective function. Compared to other models of capacity planning that generally contain a relatively a short list of micro-level factors in modelling, the proposed model contains all required macro- and micro-level factors.

As the key contribution, this thesis highlights the key drivers of capacity planning and demand modeling of supply-demand equilibrium and their relationships in the Australian flight domestic market. As a part of results, there is a bilateral relation among the elements of capacity decisions and passenger demand. The results statistically differentiate the airlines' policies of capacity planning across the different markets. The results suggest that a higher demand for flights primarily results in increased flight frequency rather than increased aircraft size or load factor. The load factor is identified to be an insignificant variable in capacity planning of the airlines. Competition between airlines, participation of low-cost carriers, and jet fuel expenses are thought to influence airlines' capacity decisions, albeit differently across the given markets. Interestingly, jet fuel cost inflations stimulate the flight demand in the short-haul market as well as the routes linking the major cities to the industrial ones. The socio-economic parameters of population and employment rates affect the flight demand in the different markets in different ways. The findings indicate the airlines' capacity decisions influence flight delays. The results indicate that more frequent flights and larger aircrafts together are associated with more flight delays. Route congestion is caused by more flights, albeit to a higher degree for low-cost carriers. Jet fuel cost inflation is expected to cause flight delays, but more for the legacy airlines than low-cost carriers.

From the results of the optimisation model, for a given period, December 2015, the optimum solutions of 3SLS and MLE indicate, respectively, a 1.72% and 0.66% improvement on the flight demand compared to the reported actual plan for the airlines. The estimated MSE of the MLE model is smaller than that of 3SLS; however, estimated coefficients of 3SLS are statistically more significant than those of MLE, resulting in more practical results in the optimisation section. The proposed model and findings of this thesis can potentially be applied by airlines as well as policy makers to fleet planning and airport infrastructure development projects in different airports and hub-and-spoke networks across the globe. The proposed optimisation model may be enhanced by using the theory of full equilibrium to develop the optimisation model through adding the factors of the other transportation modes. Due to the data limitation, airfare was only applied as an exogenous parameter in the passenger demand equation of the optimisation model. Airfare can potentially be upgraded to become a key variable of airline capacity planning under the supply-demand equilibrium. In future research, the data of individual airlines can be applied separately at the route level. With the airline dimension in modelling, further explorations can be done on the airline's policies and performance of capacity planning in different markets. The proposed model can potentially be applied to other airports and hub-and-spoke networks across the globe which it surely leads to further explorations about the airlines' policies and capacity planning as well as the demand modelling under the supply-demand equilibrium.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Business IT and Logistics
Subjects Logistics and Supply Chain Management
Keyword(s) aviation
capacity planning
supply-demand equilibrium
econometric analysis
optimization
Versions
Version Filter Type
Access Statistics: 71 Abstract Views, 23 File Downloads  -  Detailed Statistics
Created: Mon, 30 Mar 2020, 13:01:49 EST by Keely Chapman
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us