This approach permits a precise distinction between fuel burned to provide a given level of service and fuel burned as a result of inefficiencies, such as the use of older technology, circuitous routing, or taxiing with two engines instead of one.

Prior attempts to compare the fuel efficiency or environmental performance of airlines suffered from a number of limitations, the ICCT authors suggested:

• lack of transparent information about the data, which were frequently proprietary, and analytical techniques (e.g., “black box” modeling without detailed methodological description);

• failure to use real-world airline fuel consumption data for validation;

• simplistic metrics, such as fuel per passenger mile, that inappropriately penalize certain carriers;

• inability to distinguish productive from nonproductive (circuitous) passenger miles; and

• narrow coverage range of airline types, which ought to include low-cost carriers and regional affiliates of larger airlines.

The new methodology improves upon those previous efforts in four fundamental ways, the authors said:

1. It uses airline-reported fuel consumption data, rather than modeled estimates, to account fully for all the ways in which airlines can reduce fuel burn (e.g., aircraft technology or operational practices).

2. It develops an efficiency metric that recognizes that airlines burn fuel to provide both mobility (measured in terms of passenger miles traveled) and access (frequency of service and number of airports served), allowing an equitable comparison between airlines.

3. The efficiency metric distinguishes productive from nonproductive miles flown by identifying those airlines that operate particularly circuitous routes.

4. It attributes the transport service provided by and fuel consumption of affiliate carriers to mainline carriers in order to enable comprehensive comparisons across carriers’ full business operations.

Commercial passenger aircraft efficiency is defined as transport of people (the desired outcome) compared with the amount of fuel used (the required input). While gauging fuel use is relatively straightforward, measuring airline transport service is less so because of the need to account for airline business models, mobility, access, and circuity. As noted in the introduction, airlines offer their passengers both mobility and access. This study accounts for airlines that utilize fewer (or more) stops to transport a passenger from origin to destination by evaluating fuel use relative to both revenue passenger miles (RPM) and the number of departures—RPMs being a measure of the mobility provided by an airline and departures serving as a proxy for accessibility.

In general terms, the more RPMs an airline provides per unit of fuel, the more efficient the airline. However, travelers’ ability to avail themselves of an airline’s RPMs is also important, with each airline striking a unique balance between getting passengers to their final destination as directly as possible with the need to provide access to all its potential customers. Carrying more, flying further, and offering more flights—either by serving more airports or flying more frequently—all increase fuel burn. A robust methodology credits airlines for fuel consumed to supply both RPMs and departures.

At the same time, while the intention is to credit fuel used to enhance access, there needs to be a means of discounting fuel used for overly circuitous flights. Circuity is the ratio of the actual distance traveled to the intended distance of travel (i.e., the great circle distance from origin to destination); by definition, circuity is always equal to or greater than 1.

...The greater the value of circuity, the more nonproductive miles are traveled by an airline’s customers. Airlines deviate from great circle distances for an origin-destination pair for various reasons, including weather and air traffic control restrictions. However, an airline’s network structure also influences its level of circuity and hence the number of nonproductive miles traveled by passengers.

—Zeinali et al.

Among the findings:

• Alaska Airlines led all fifteen mainline carriers operating in the US in 2010, followed closely by Spirit Airlines and Hawaiian Airlines—all three being relatively small carriers serving geographically limited markets.

• Continental airlines, in fourth place, was the most fuel-efficient full-service legacy carrier (established prior to deregulation) while Southwest Airlines (fifth) was the most efficient carrier operating on a point-to-point rather than a hub-and-spoke business model.

• United Airlines (eighth), with a fuel efficiency score of 1.00, was equal to the average for 2010 US domestic operations.

• The least efficient airline in this ranking, Allegiant Air, also happened to have the most profitable US domestic operations during the 2009 to 2011 period.

Many, although not all, of the carriers with worse fuel efficiency than the industry average were, or subsequently have been, the subject of merger activity, including Delta Air Lines (11^th), US Airways (12^th), Airtran Airways (13^th), and American Airlines (14th).

In-use fuel efficiency score (y-axis) versus average aircraft age (in years, x-axis). Zeinali et al. Click to enlarge.

The report attributes roughly a third of the variation in fuel efficiency to the level of investment in technologies such as winglets and high-bypass ratio engines, which is related to aircraft age. Factors such as seating density and percent occupancy, ground operations practices such as single-engine taxiing, and how well an airline matches aircraft capability with need on any given route account for the remainder.

In addition to overall airline fuel efficiency, the study analyzed the ten most important city-to-city routes in the continental US; this analysis relied on an aircraft performance model to estimate fuel burn. On those routes the estimated difference between the most- and least-efficient airlines was even greater than overall, ranging from 9% (San Francisco–San Diego) to 87% (New York–Chicago) because of aircraft technology, load factor, seating density, and route circuity.

Shorter trips—e.g., North-South travel along the coasts—are significantly more fuel intensive on a passenger-mile basis than transcontinental flights. The authors attributed this general finding in part to the large amount of fuel consumed during takeoff and to the use of smaller, less efficient aircraft such as regional jets on shorter routes.

In many cases, because of variations in aircraft choice, seating density, route circuity, and such, an airline offering efficient flights between a major city pair may not rank well at the network level. The results highlight the need for better route- and even flight-specific data for consumers wishing to select less carbon-intensive flights for a specific trip, the authors suggested.

Overall, the authors drew four major conclusions:

1. Within the mature, competitive US aviation market, there remains a large gap in fuel efficiency among airlines flying on domestic routes even at today’s high fuel prices. Fuel prices alone may not be a sufficient driver of in-service efficiency across all airlines.

2. Although technology utilization is important, operational practices vary significantly from airline to airline, impacting in-service fuel efficiency. To address the energy and environmental effects of aviation, policies that facilitate and encourage efficient operations as well as technologies should be considered.

3. the most efficient airline for a given itinerary varies by city pair, with the top performer on one route not necessarily being efficient on another. Route-specific data are required for consumers to make more environmentally responsible decisions.

4. Accurate, transparent data are the cornerstone for assessing the efficiency of airlines. Reporting requirements for more detailed data on aircraft fuel consumption would go further in supporting any future policies to improve efficiency.

The ICCT plans to update the report annually, with numbers for 2011 slated for release later this year.

Getting the methodology right and establishing this baseline was the hardest part. Because we rely strictly on publicly reported data, the updates will be pretty straightforward.

—Dan Rutherford, co-author

Mazyar Zeinali, a report co-author, noted that the International Civil Aviation Organization (ICAO), which sets technical standards for aircraft safety and emissions, could address both technical and operational inefficiencies by establishing an aircraft CO₂ emission standard and throwing its weight behind some type of market-based scheme to impose a price on carbon emissions. The ICAO General Assembly will convene in Montreal beginning on 24 September 2013.

NEXTOR study. The NEXTOR team published a report, sponsored by the ICCT, in November 2012 describing their methodology and applying it to 15 operators in the US.

Multiple approaches are employed, based upon different models of the relationship between airline fuel consumption and production. In addition, we explicitly recognize—to our knowledge for the first time—that affiliations between these large jet operators and regional carriers must be taken into account when assessing the fuel efficiency of large jet operators. We also measure airline fuel efficiency with respect to passenger trips, by using a passenger origin-destination (O-D) based airline output metric as an alternative to the standard passenger-mile metric, which ignores the effect of circuitous routings.

By developing different models, considering large jet operators by themselves or in combination with their regional affiliates, and measuring carrier output in different ways, we obtain a range of fuel efficiency metrics and rankings. While we do not want to choose one best answer out of these, we do provide in this report a detailed exposition of alternative methods for measuring and ranking airline fuel efficiency, and the assumptions underlying the different methodologies. We compare the results from the different approaches, in an effort to provide additional insights into the factors that cause the ranking difference. Our ultimate aim in these efforts is to develop a robust, transparent, airline fuel efficiency assessment framework that can eventually be extended to other airlines around the globe as long as equivalent data are available.

—Zou et al. (2012)

In the work, they used ratio-based (Fuel/Revenue Passenger Origin-Destination Miles, deterministic, and stochastic frontier approaches—all explained in detail in their report.

Overall, their findings were that:

1. Airline fuel consumption is highly correlated with, and largely explained by, the amount of revenue passenger miles and flight departures it produces;

2. Depending on the methodology applied, average airline fuel efficiency for the year 2010 is 9-20% less than that of the most efficient carrier, while the least efficient carriers are 25-42% less efficient than the industry leaders;

3. Regional carriers have two opposing effects on fuel efficiency of mainline airlines: higher fuel per revenue passenger mile but improved accessibility provision; and

4. The net effect of routing circuity on fuel efficiency is small.

Resources

• Mazyar Zeinali, Daniel Rutherford, Irene Kwan, an Anastasia Kharina (2013) US Domestic Airline Fuel Efficiency Ranking 2010 (ICCT)

• Bo Zou, Matthew Elke, Mark Hansen (2012) Evaluating Air Carrier Fuel Efficiency and CO₂ Emissions in the US Airline Industry (original NEXTOR report)

• Bo Zou, Matthew Elke, Mark Hansen (2013) Evaluating Air Carrier Fuel Efficiency in the US Airline Industry (Tenth USA/Europe Air Traffic Management Research and Development Seminar (ATM2013))