Search

The Federal Aviation Administration (FAA) is continuously assessing the operational performance of the National Airspace System (NAS), where they analyze trends in the aviation industry to help develop strategies for a more efficient air transportation system. To measure the performance of various elements of the aviation system, the FAA and the International Civil Aviation Organization (ICAO) developed nineteen key performance indicators (KPIs). This dissertation contains three research studies, each written in journal format, addressing select KPIs. These studies aim at answering questions that help understand and improve different aspects of airport operational efficiency. In the first study, we model the flight times within the terminal airspace and compare our results with the baseline methodology that the FAA uses for benchmarking. In the second study, we analyze the efficiency of deicing operations at Chicago O'Hare (ORD) by developing an algorithm that analyzes radar data. We also use a simulation model to calculate potential improvements in the deicing operations. Lastly, we present our results of a clustering analysis surrounding the response of airports to demand and capacity changes during the COVID-19 pandemic. The findings of these studies add to literature by providing a methodology that predicts travel times within the last 100 nautical miles with greater accuracy, by providing deicing times per aircraft type, and by providing insight into factors related to airport response to shock events. These findings will be useful for air traffic management decision makers in addition to other researchers in related future studies and airport simulations. Doctor of Philosophy The Federal Aviation Administration (FAA) is the transportation agency that regulates all aspects of civil aviation in the United States. The FAA is continuously analyzing trends in the aviation industry to help develop a more efficient air transportation system. They measure the performance of various elements of the aviation system. For example, there are indicators focused on the departure phase of flights measuring departure punctuality and additional time in taxi-out. On the arrivals side, there are indicators that measure the additional time spent in the last 100 nautical miles of flight. Additionally, there are indicators that measure the performance of the airport as a whole such as the peak capacity and the peak throughput. This dissertation contains three research studies, each one aims at answering questions that help understand and improve a different aspect of airport operational efficiency. The first study is focused on arrivals where we model the flight times within the last 100 nautical miles of flight. Our model incorporated factors such as wind and weather conditions to predict flight times within the last 100 nautical miles with greater accuracy than the baseline methodology that the FAA currently uses. The resulting more accurate benchmarks are important in helping decision makers, such as airport managers, understand the factors causing arrival delays. In the second study, we analyze the efficiency of deicing operations which can be a major source of departure delays during winter weather. We use radar data at Chicago O'Hare airport to analyze real life operations. We developed a simulation model that allowed us to recreate actual scenarios and run what-if scenarios to estimate potential improvements in the process. Our results showed potential savings of 25% in time spent in the deicing system if the airport changed their queueing style towards a first come first served rather than leaving it for the airlines to have their separate areas. Lastly, we present an analysis of the response of airports to demand and capacity changes during the COVID-19 pandemic. In this last study, we group airports by the changes in their throughput and capacity during two time periods. The first part of the study compares airports operations during 2019 to the pandemic during the "shock event" in 2020. The second part compares the changes in airports operations during 2020 with the "recovery" time period using data from 2021. This analysis showed which airports reacted similarly during the shock and recovery. It also showed the relationship between airport response and factors such as what kind of airlines use the airport, airport hub size, being located in a multi-airport city, percentage of cargo operations. The results of this study can help in understanding airport resilience based on known airport characteristics, this is particularly useful for predicting airport response to future disruptive events.

This dissertation looks at the relationship between airport size (e.g. small, medium, large) and air service continuity and operational performance. It consists of three studies, each written in journal format. The first study analyzes the markets served pre- and post-recession while focusing on the operational strategies adopted by the top Major Carriers and Low-Cost Carriers (LCCs) in the United States. Findings show that LCCs have outpaced major carriers in terms of expanding their network and the number of markets served. During the same time, major carriers have gained a greater flight share in the markets they already serve. Post-recession, LCCs have shown preference to competing with major carriers over other LCCs. The second study investigates the declining service levels at small airports compared to large-hub airports, which continue to benefit from higher levels of service and increased airline presence. Using a fixed-effects conditional logistic regression, this study looked at factors contributing to service loss in region-to-region markets serving small communities between 2007 and 2013. Results show that 1) markets affected by a merger are indeed at a higher risk of losing service; 2) markets that are operated by a fuel-intensive, small-aircraft fleet have a higher chance to be discontinued and 3) an increased number of competitors greatly reduces potential market service loss. The third and final study proposes a new methodology to calculate original delay and propagated delays using combined aviation operational datasets that provide detailed flight information and causal factors behind delays. In addition to calculating original and propagated delay for the month of July of 2018, this study differentiated between original delays that occur during the turnaround phase, taxiing phase and en-route and incorporates causal factor information to identify the true source behind propagated delay. Two fixed-effects linear regression models were introduced that predict Total Propagated Delay and the share of propagated delay given an airport's ability to absorb upstream delay during the turnaround phase. Results show that most delay propagation chains originate at large-hub airports and are mostly concentrated at airports within the same geographical area. However, delays originating at large-hub airports were found to be the quickest to recover (i.e. least number of downstream flight legs affected) and large-hub airports have a higher ability to absorb delay at the turnaround phase compared to smaller airports given the significantly higher schedule buffer time airlines plan at large-hub airports. Doctor of Philosophy The changing nature of the air service industry is dependent on several key factors, including but not limited to the major and low-cost airlines, the frequency of service at different sized-airports and the operational performance of the airports in the system. Each airport can be classified by size based on the annual number of enplanements. This dissertation looks at the relationship between airport size (e.g. small, medium, large), service continuity and operational performance. It consists of three studies, each written in journal format. Over the past two decades, the U.S. air transportation network witnessed several economic downturns forcing airlines to shift their operational strategies, cease service or merge with an airline counterpart. The first study analyzes routes served before and after the recession by exploring the presence of major and low-cost carriers in these markets to understand how several economic downturns have influenced the operating strategy of airlines in the US. While Low-cost carriers focused on expanding their network and offering service in an increased number of new routes, major carriers increased their presence in the markets in which they already serve. Furthermore, after the recession, low-cost carriers chose to increasingly compete with major carriers over their low-cost counterparts. The second study explored the factors that can potentially contribute to the loss of service in routes serving small communities. While airlines continue to compete on the most profitable routes, small airports recently suffered from reduced service levels and in some instance service discontinuity. Results show that 1) routes that were once served by two airlines that merged are at a higher risk of losing service; 2) routes that are operated by a fuel-intensive small aircraft fleet have a higher chance to be discontinued and 3) an increased presence of airlines competing in a route greatly reduces potential service loss. In addition to evaluating service continuity, the third and final study looks at flight delays across the US and dives into the effect of airport size on propagated delay. Delays on a flight can be caused by inefficiencies and capacity restrictions at airports and may also be the result of delay that happen earlier in the day and that propagates to multiple flights downstream that share the same resources. That is, a delay can affect multiple flights whenever these flights are all operated by the same aircraft equipment. Costing the air transportation network billions of dollars annually, the third study examines the original and propagated delays at US airports by collecting data from multiple sources to incorporate the original source and cause of delay. Results show that most delay originates at large-hub airports and are mostly concentrated at airports within the same geographical area. However, delays originating at large-hub airports were found to be the quickest to recover and large-hub airports have a higher ability to absorb delay at the turn compared to smaller airports as airlines allocate additional minutes of schedule padding at large-hub airports.