BTS Outline of Statistical Highlights and Visuals
C-SPAN Washington Journal: How U.S. Airlines Perform

HTML | PDF | PPT

April 13, 2012

The big picture

The performance of U.S. airlines has been the subject of public interest because of the growing number of people who fly. In 2011, more than 800 million passengers flew on U.S. airlines and on foreign airlines serving the U.S. At times in the past, U.S. airlines' performance has been the subject of criticism because of low-on-time rates and numerous lengthy tarmac times.

On recent months, airlines have been posting record high on-time arrival rates, record low cancellation rates and have virtually eliminated long tarmac times. Airline performance is strongly dependent on weather and traffic volume, as well as each airlines individual operation and the operation of the air traffic control and airport system.

Slides following the cover and the summary of recently released data:

1. U.S. Air Travel On-Time Performance

• Cover page

2. On-time, cancellation, tarmac and fuel cost data for February 2012

• February, with 86.2 percent of flights arriving on time, had highest on-time performance for any February and sixth highest for any month in 18 years with comparable records.
• February cancellation rate of 1.0 percent was the lowest for any February.
• No tarmac times of more than 3 hours in February, fifth month in last 13 with no lengthy domestic tarmac times.
• Airline fuel costs continued to rise in February – at $2.98/gal up 7.6 percent from February 2011

Visual: Headline Page
(Bulleted text with results of most recent BTS data releases prior to April 13, 2012)

3. Air Passenger Travel Grew 61% from 1990 to 2011

• U.S. airlines and airports have faced increasing demands
• Passenger numbers have increased steadily except for declines following the recession of 2000 and 9/11 and the recession of 2007 to 2009.

Visual: Demand on U.S. Aviation System Continues to Increase:
(Line graph showing the number of airline passengers within the U.S. and on flights to and from the U.S. from 1990 through 2011)

4. 2 million passengers flew on 25-30,000 flights a day in 2011

• 2 out of 5 Americans fly at least once a year
• 194 airlines provide scheduled service to 604 airports in the U.S.
• Dour of 5 flights were on time in 2011

Visual: Air Passenger Travel Statistics
(Bulleted text with air passenger travel facts)

5. The largest airports show growth in passenger traffic over two decades

• Four airports handled more passengers in 2011 than the top airport in 1990
• Atlanta, which handled 92 percent more passengers in 2011 than 1990, replaced Chicago O'Hare as the busiest airport
• Las Vegas and Houston Bush replaced Miami and New York LaGuardia in the top 10.

Visual: In 2011, the Top 10 Airports Accounted for Almost One-Third of All Air Passenger Travel
(Table showing top 10 U.S. airports in 1990 and 2011)

6. Airline fuel costs have been rising in recent months

• Fuel costs make up 33 percent of airline expenses.
• Fuel prices have been rising. Airlines paid 21 cents per gallon more in February 2012 than February 2011.
• Fuel costs tend to rise in spring and early summer.

Visual: U.S. Airlines Spent $47 Billion on 16.4 Billion Gallons of Fuel in 2011
(Bar graph showing airline fuel cost per gallon by month)

7. Airline performance is measured against published schedules.

• An on-time arrival must arrive at the gate within 15 minutes of scheduled arrival time.
• An on-time departure must depart the gate within 15 minutes of scheduled departure time.

Visual: How We Measure On-Time Performance
(Bullets explaining how airline on-time performance is measured)

8. On-time data must be reported by carriers that meet the required revenue threshold.

• Airlines with 1 percent of the scheduled service domestic revenue for all airlines are required to report.
• In 2012, 14 airlines meet the revenue threshold and one carrier reports voluntarily.
• As few as 10 carriers were required to report in 2002 and as many as 20 were required to report in 2007

Visual: Who Reports On-Time Performance
(Bullets explaining how airline on-time performance is reported)

9. Airline on-time arrival performance depends greatly on the weather and the number of flights.

• Airline on-time arrival percentage reached a peak in 2002 and 2003 when airlines reduced flights after 9/11 and the 2000-2001 recession.
• Performance improved again after 2007 as airlines cut capacity in response to rising fuel costs.
• Performance reached record levels in late 2011 and early 2012 during mild winter weather.

Visual: On-Time Performance 1995-2011
(Line graph showing annual on-time arrival percentage from 1995 through 2011)

10. On-time arrival performance has reached record highs in recent months

• On-time arrival percentage was record high in December, January and February for each of those months since 1995
• Three of the top 12 months since 1995 took place in the last five months – October, November and February.

Visual: Airline On-Time Arrival Performance by Month Jan 2011-Feb 2012
(Bar graph showing monthly on-time arrival percentage)

11. The cancellation rate reached record lows in recent months

• Cancellation rates were at record lows in December, January and February for each of those months since 1995
• Four of the lowest 26 months since 1995 took place in the last five months – October, November, December and February.

Visual: Airline Flight Cancellation Rate by Month Jan 2011-Feb 2012
(Bar graph showing monthly cancellation percentage)

12. Data was reported in 2011 by more airlines and a different mix of airlines than in 2000.

• 10 airlines reported on-time data in 2000, virtually all major network airlines with Continental reporting the highest on-time performance and United the lowest.
• 16 airlines reported in 2011, including low-cost and regional airlines, with Hawaiian reporting the highest on-time performance and JetBlue the lowest.
• Hawaiian's top ranking shows the impact of weather on airline performance.

Visual: Airline On-Time Arrival Rankings – Top 5 and Bottom 5 2000 vs. 2011
(Table listing the top 5 and bottom 5 airlines for on-time arrival performance in 2000 and 2011)

13. Airport on-time arrival rankings show similar patterns in 2000 and 2011.

• Three of the same five airports were in the bottom five in 2000 and 2011 – Boston, New York LaGuardia and San Francisco.
• The bottom five airports in both years show the impact of congested airspace and poor weather in the northeast compared to other parts of the country, as well as the San Francisco fog conditions.
• The top five airports have generally been located in the west in recent years.

Visual: Airport On-Time Arrival Rankings – Top 5 and Bottom 5 2000 vs. 2011
(Table listing the top 5 and bottom 5 airports for on-time arrival performance in 2000 and 2011)

14. On-time performance declines as the day progresses even at the best airport but not as much as at the bottom airport.

• Newark airport saw a much steeper decline during the day and less of a recovery in the evening.
• Salt Lake City's performance dropped to Newark's level in late morning but remained above it the rest of the day.
• Newark's on-time arrival percentage was above 80 percent in the morning but dropped below 60 percent later in the day.

Visual: On-Time Performance Declines During the Day
(Line graph comparing the on-time arrival performance by time of day in 2011 for Salt Lake City, the highest performing airport in 2011, with Newark, the lowest performing airport)

15. On-time performance in the New York-New Jersey area is well below the national average.

• The national on-time arrival percentage in 2011 was 80.1 percent if New York-New Jersey airports were not included.
• The on-time arrival percentage in 2011 at Newark, New York JFK and New York LaGuardia was 71.2 percent..
• Including the New York-New Jersey airports, the national on-time arrival percentage in 2011 was 79.6 percent.

Visual: On-Time Arrival Performance in New York Area Compared to Nation
(Bar graph comparing 2011 on-time arrival performance at the three New York-New jersey airports with the national average)

16. Lengthy tarmac times have virtually disappeared in recent months

• No tarmac times of more than three hours in the last three months.
• In the last 14 months, there have been five months with no domestic flights with three hour tarmac times and two more with only one.
• BTS only has data on international flights at U.S. airports since August. International flights may have problems that domestic flights don't such as waiting for customs gates to be available.

Visual: For 5 of Last 13 Months, No Domestic Flight On Tarmac Longer than 3 Hours
(Bar graph showing tarmac times of more than 3 hours by month since October 2010)

17. Far few lengthy tarmac times have been reported since the DOT enforcement rule took effect in April 2010

• From October 2008 through April 2010, there were eight months with 50 flights with tarmac times of more than three hours, including on with more than 250 and two others with more than 100.
• Since May 2010, there was one month with 16 three-hour tarmac times, one with 14, one with seven and no others with more than four.
• Since May 2010, there have been six months with no flights with three-hour tarmac times.

Visual: Tarmac Delays Before and After April 2010 Tarmac Delay Rule
(Bar chart showing the number of 3-hour tarmac times by month from Oct 2008 through Jan 2012. A longer view of the previous slide.)

18. Tarmac times and cancellation rates increase at the same time in some months but in other months they do not

• Cancellation rate has generally risen at the same time that lengthy tarmac times increase, such as in late 2008, early 2009, February 2010 and early summer 2011.
• The cancellation rate spiked in February 2010 and February 2011 when airline operations were hindered by snowstorms.
• Lengthy tarmac times spiked in June 2009 in the face of summer thunderstorms.

Visual: Tarmac Delay Rule and Cancellation Rate
(Line graph with one line showing the number of three-hour tarmac times by month from Oct 2008 through Jan 2012 and a second line showing the cancellation rate.)

19. Despite record low cancellation rates in recent months, millions of passengers are still impacted.

• As many as an estimated 1.5 million passengers were impacted by cancellations during the snowstorm-plagued month of February 2011.
• More than 1 million passengers were impacted in August 2011
• In recent months, the number of impacted passengers has been reduced by the record low cancellation rates.

Visual: Over 12 Months, 104,000 Flights Canceled, 7.5 Million Passengers Impacted
(Line graph comparing cancellation rate by month from Feb 2011 through Jan 2012 with the estimated number of passengers affected.)

20. Causes of delay reported by the airlines are distributed among several categories.

• Aircraft Arriving Late (40%) was the single largest category of delay in 2011, showing the ripple effect of previous delays.
• Delay attributed to the airlines (30%) was the second largest category.
• Air traffic control and airports that make up the National Aviation System were the cause of one-fourth of delay minutes (25%).

Visual: Causes of Flight Delays, 2011
(Pie chart showing the percentage of minutes of delay attributed to each category of cause of delay in 2011)

21. The percentage of delay minutes attributed to each category has varied since 2003.

• Aircraft arriving late on the previous leg of a flight has become the single biggest cause of delays.
• Delay attributed to the airlines became the second largest category in 2009.
• Air traffic control and airports was the top clause of delay in 2003 but has been the third largest since 2009.

Visual: Trends in the Causes of Flight Delays
(Line graph comparing the percentage of minutes of delay attributed to each category of cause of delay annually since 2003)

22. The impact of weather on airline operations ranges from extreme weather that prevents flying to milder weather that slows flights.

• BTS uses the weather data reported in various categories to compute the full scope of weather-related flight delays.
• Weather share of delay minutes is computed by BTS using airline reports on Extreme Weather and weather reported in the Air Traffic Control and Airports category.
• Using the reported categories, BTS computes the share of weather-related delay in the Late Arriving Aircraft category.

Visual: Weather Accounts for Almost 40% of Delays
(Bar graph comparing the percentage of minutes of delay attributed to weather annually since 2003)

23. Delays and cancellations have a seasonal pattern with rates varying from month to month depending on the number of flights and the weather.

• The percentage of flights arriving late peaks in summer. This number is for flights arriving at their destination and does not include cancellations or diversions.
• Cancellations tend to rise during the winter months in the face of severe weather.
• The cancellation rate also rises in September, a month when hurricanes are frequent.

Visual: Late Arriving Flights and Cancellations by Month, 1995-2011
(Line graph comparing the average percentage of late arriving flights to the average cancellation rate for each of the 12 months since 1995)

24. The cancellation rate usually increases in winter.

• Cancellations rates are very seasonal, spiking in winter when snowstorms hinder flying.
• Airline mergers also lead to cancellations as the merging airlines attempt to combine their systems.
• The low cancellation rates in 2011-2012 reflect the mild winter weather.

Visual: Winter Weather Contributes to Higher Flight Cancellation Rates
(Line graph showing the cancellation rate by month from 2003 through January 2012)

25. Baggage fees have become a growing source of revenue for airlines since 2007.

• Airline revenue from baggage fees increased each year from 2007 through 2010.
• Nine-month numbers for 2011 show a slowing of the increase from 2010.
• Airlines collected almost $3.4 billion from baggage fees in 2010 compared to $464 million in 2007.

Visual: Baggage Fee Revenue ($ millions)
(Bar graph with blue bars showing airline revenue from baggage fees annually from 2007 through 2010 and red bars showing revenue from January through September of each year from 2007 through 2011.)

26. Thanksgiving 2011 showed the importance of the number of flights and weather to on-time performance.

• The peak passenger travel days for the Thanksgiving holiday period are the Tuesday and Wednesday before the holiday and the Sunday and Monday afterwards. With the exception of Tuesday Nov. 22 (77.5 percent), the on-time performance of the other peak travel days varied little from the monthly average of 85.3 percent in November 2011.
• With snowstorms in the southeast, on-time performance at Atlanta was well below the national average on the Tuesday before the holiday, 67.4 percent.
• Recognizing that fewer passengers travel on Thanksgiving and the day afterwards, airlines reduced their daily flight operations by 26 percent compared to the monthly daily average. As a result, Thanksgiving's on-time performance was 93.8 percent and the day after result was 95.8 percent.

Visual: Flying during Thanksgiving Holiday Period
(Line graph showing on-time arrival percentage by day during November 2011)

27. BTS has expanded the collecting of tarmac data in stages.

• From 1997 through September 2008, BTS collected the taxi-in and taxi-out times for scheduled service domestic carriers that reported on-time data.
• Beginning October 2008, BTS began collecting tarmac data on canceled and diverted flights from the carriers reporting on-time data.
• In August 2011, all airlines serving the United States were required to report any tarmac time of more than three hours at a U.S. airport.

Visual: Bureau of Transportation Statistics Tarmac Time Collection
(Text presenting the chronology of BTS tarmac time collection.)

28. BTS began collecting cause of delay and cancellation data in 2003.

• The broad categories for reporting the causes were recommended by a task force that included government and industry representatives..
• A complete list of the examples for each category can be found in Section IX of Technical Directive #21.

Visual: Categories for Reporting Causes of Flight Delays
(Text presenting the categories for reporting of causes of flight delays.)