CODA Digest All- Causes Delay and Cancellations to Air Transport in Europe Quarter

CODA Digest All-Causes Delay and Cancellations to Air Transport in Europe Quarter 4 2014

FOREWORD This report gives an overview of the delay situation in the European Civil Aviation Conference Area. This report has been prepared by the Central Office for Delay Analysis (CODA), a service of EUROCONTROL. It is based on the EUROCONTROL CODA database which contains delay data provided directly by airlines. The analysis of operational cancellations is based on airport provided data under Commission Regulation N 691/2010 further information is available on the CODA portal at http://www.eurocontrol.int/articles/ir691-data-collection-process. The report consists of an overview of the reporting period, a summary of the main delay effects, and a series of charts and graphics, which illustrate the main characteristics of the reporting period. A glossary of terms and abbreviations used throughout the report is given in Annex Glossary of Terms and Abbreviations. Airline data from the CODA database contains real recorded delays provided to CODA by airlines and is based on the difference between scheduled time of departure and actual off block time. Air Traffic Flow Management (ATFM) delays reported by airlines may be lower than the Network Manager (NM) calculated ATFM delays due to difference in methods: ATFM delays of NM are the (flight) planned delays ; the airlines report the actual experienced ATFM delay on departure. For instance, a flight with an ATFM delay may also have a handling delay absorbed within the ATFM delay. For the airline, a part of this delay is the ATFM delay and the rest is the handling delay. This report discusses delays from all-causes; for information regarding ATFCM delays see the complementary reports available at http://www.eurocontrol.int/articles/network-operations-monitoring-and-reporting Table of Contents 1. Headlines and Monthly Events Overview.... 3 2. CODA Operational Flight Cancellation Analysis... 5 3. CODA Scheduling Indicators... 7 4. CODA Reactionary Delay Analysis... 8 5. Average Delay per Flight for Departures... 9 6. NM versus Aircraft Operator experience of Delay... 10 7. Average Delay per Flight by Hour... 11 8. Average Delay per Delayed Flight >5 Minutes... 12 9. Distribution of All Flights by Length of Delay (Punctuality)... 13 10. Average Delay per Flight for Arrivals... 14 11. Top 20 Affected Departure Airports... 16 12. Top 20 Affected Arrival Airports... 17 13. Top 20 Most Delayed Airport Pairs... 18 14. Year-on-Year Trends in All-Causes Delay Indicators... 19 A. Glossary of Terms and Abbreviations... 20 B. CODA Delay Groupings... 21 C. Correlation between IATA Codes and the NM Regulation Codes... 22 D. Standard IATA Delay Codes (AHM730)... 23 E. Standard IATA Delay Sub-Codes (AHM731)... 25 F. CODA Coverage Of Commercial Flights in the ECAC Region Q4 2014... 26 2015 European Organisation for the Safety of Air Navigation (EUROCONTROL) This document is published by EUROCONTROL for information purposes. It may be copied in whole or in part provided that EUROCONTROL is mentioned as the source and to the extent justified by the non-commercial use (not for sale). The information in this document may not be modified without prior written permission from EUROCONTROL. The use of this document is at the user s sole risk and responsibility. EUROCONTROL expressly disclaims any and all warranties with respect to any content within the document, express or implied. Central Office for Delay Analysis EUROCONTROL 96 Rue de la Fusée B - 1130 Brussels The Central Office for Delay Analysis (CODA) is ISO 9001:2008 certified. Tel: + 32-2 729 33 91 Fax : + 32-2 729 90 04 E-Mail : coda@eurocontrol.int Web Site: http://www.eurocontrol.int/coda Date of publication of this issue: 18/02/2015 EUROCONTROL 1 CODA

1. Headlines and Monthly Events Overview. Q4 2014 Summary. As reported in the Monthly Network Operations Report during the fourth quarter of 2014 (Q4 2014) there were 1.1% more flights per day in the EUROCONTROL Statistical Reference Area (ESRA08) than in Q4 2013 (Figure 1). During the quarter disruption in Ukraine continued, Libyan and Syrian airspace closures alongside the opening of the KFOR sector saw changes in traffic flows. Airline strikes, ATC industrial action, a shifting jetstream and winter weather conditions were observed during the quarter. ATFM delay improved with an average delay per flight of 0.78 minutes being observed in comparison to 0.83 minutes per flight in Q4 2013. Data direct from airlines to CODA describing delays from all-causes illustrated an increase in delay in the final quarter of the year (see Figure 3) with an average delay per flight (ADM) of 9.6 minutes on departure, in comparison to 9.2 minutes per flight observed in Q4 2013. 37% of flights were delayed on departure (>=5 minutes) this was an increase of 1.6 points. The share of reactionary delay was 44% compared to 45% of delay minutes reported in Q4 2013. Regarding arrival delay the average delay per flight, this remained stable at 9.2 minutes per flight compared to 9.3 minutes in Q4 2013. Q4 2014 Detailed Monthly Summary. October 2014 data from airlines describing all-causes delay showed an average delay per delayed flight (ADD) of 25 minutes. 36% of flights were delayed on departure (>=5 minutes), this was an increase of 2.3 percentage points when compared to the same month in 2013. On arrival, 34% of flights were delayed and showed an average delay per delayed flight of 27 minutes. Amsterdam, Frankfurt, London Heathrow, Berlin Tegel and Paris De Gaulle airports were particularly affected by seasonal weather (low visibility and thunderstorms) in October. Industrial action at Germanwings on 15 October and at Lufthansa on 20 and 21 October, resulted in more than 1500 flights being cancelled. In November 2014 the average delay per delayed flight (ADD) was 22 minutes, a fall of 2.1 minutes when compared to November 2013. 32% of flights were delayed on departure (PDF >=5 minutes). On arrival, 29% of flights were delayed and showed an average delay per delayed flight of 24 minutes compared to 25 minutes in November 2013. London Heathrow, Amsterdam, Zurich, Frankfurt, Brussels, Berlin Tegel and Tenerife South airports were particularly affected by seasonal weather (strong winds and thunderstorms). The jetstream position saw traffic shifting over Lisbon with resultant capacity delays being observed. There was also industrial action in Greece between 26 & 27 November. December 2014 all-causes delay data showed an average delay per delayed flight (ADD) of 28 minutes, a decrease of 2.9 minutes when compared to December 2013. 43% of flights were delayed on departure (PDF >=5 minutes). On arrival, 39% of flights were delayed and showed an average delay per delayed flight of 30 minutes compared to 32 minutes in December 2013. Industrial action at Lufthansa (on 1, 2 and 4 December) and at TAP Air Portugal (2 December) resulted the cancellation of approximately 1,600 flights. Industrial action also took place in Belgium, resulting in approximately 300 flights being cancelled on 8 December and 800 flights on 15 December. Industrial action took place in Italy on the 12 December where 1,000 flights were cancelled as consequence of the action. Seasonal weather affected Amsterdam, London Heathrow, Istanbul Ataturk, Istanbul Sabiha Gokcen and Frankfurt airports. A flight data server failure occurred at London ACC on 12 December with London Heathrow being impacted the main effect being delays. Figure 1. Total Departures per day in ESRA08 Figure 2. Long Departure Delay >60 minutes In Q4 2014 traffic increased: there were 1.1% more flights per day in the EUROCONTROL Statistical Reference Area (ESRA08) than in Q4 2013. In comparison to Q4 2013, the fourth quarter of 2014 saw small decreases in the percentage of flights delayed by greater than 60 minutes for the month (Oct 0.0, Nov -0.1 and Dec +0.3). EUROCONTROL 2 CODA

Figure 3. Delay Causes Q4 2014 vs. Q4 2013 Q4 2014 saw an increase in the average delay per flight to 9.6 minutes for all-causes delay. Further analysis of the delay reasons shows that reactionary delays increased slightly by 0.1 minutes to 4.2 minutes per flight. Airline-related delays also increased, these to 3.0 minutes per flight. ATFCM en-route delays fell to 0.3 minutes per flight in Q4 2014. Figure 4. Primary delay causes, Q4 2014 vs. Q4 2013 Figure 5. ATFCM Share of All-Causes Primary Delay. Following the deterioration in the delay situation, airline delay increased from 2.8 minutes to 3.0 minutes per flight. ATFCM delays remained stable at 1.2 minutes per flight. The ATFCM share of all-causes delay (primary delay) minutes was 22.8%. This was a 1 percentage point decrease when compared to Q4 2013. The enroute percentage share of the total primary delay minutes was 6% and a decrease of 2.1 points. EUROCONTROL 3 CODA

2. CODA Operational Flight Cancellation Analysis This section is based on data supplied by airports as per Annex IV of EC Regulation N 691/2010. These preliminary results are based on operational cancellation data supplied by 30 of the 50 European coordinated airports reporting to CODA under EC Regulation N 691/2010. Albeit based on data supplied by a restricted list (see the conditions below) of major European airports, these preliminary results already give a good indication on trends and effects of Network events like industrial actions or extreme weather. The IFR flight coverage at the individual airports included in this analysis reaches 100%. Those airports currently unable to report operational cancellations in the required data format or those not meeting all of the criteria for operational cancellations were excluded from the analysis. For the 2014 data going forward, the quality and completeness will improve. According to Annex IV of EC Regulation N 691/2010 an Operational cancellation means an arrival or departure scheduled flight to which the following conditions apply: The flight received an airport slot, and The flight was confirmed by the air carrier the day before operations and/or it was contained in the daily list of flight schedules produced by the airport operator the day before of operations, but The actual landing or take-off never occurred. This section of the CODA Digest reports on monthly, weekly and daily shares of operational cancellations of the total planned flights at day -1. Like delays, operational cancellations provide an insight into the impact of network events and associated disruption; e.g. industrial action or extreme weather events. Figure 6. Monthly Share of Operational Cancellations Jul 2013 - Sep 2014 Figure 7. Daily Share of Operational Cancellations Q4 2013 & Q4 2014 The monthly share of operational cancellations in Q4 2014 increased when compared to Q4 2013. October saw a rate of 1.7% with airline industrial actions during the month, though the definition of an operational cancellation results that only the first day (in case of strikes over multiple consecutive days) will generate operational cancellations. The highest monthly share of operational cancellations for the quarter occurred in December 2014 with multiple airline and air traffic industrial actions which resulted in a large number of cancelled flights peaking at 2.4% for the month. There was also a flight data server failure at London ACC on 12 December with London Heathrow being impacted although the effect was more of delays rather than cancellations. Figure 7 shows the percentage share of cancellations by day of the week. Cancellation rates on Mondays and Tuesdays saw decreases in comparison to Q4 2013. Also of interest in this graph is the trend of fewer cancellations at weekends, however in Q4 there were small increases over Saturday and Sunday of 0.3 & 0.2 points. With less flights scheduled during weekend days airlines may have additional spare fleet capacity and are better able to make aircraft changes quickly and maintain the operational schedule. EUROCONTROL 4 CODA

Figure 8. Average Daily Cancellation Oct 2013 - Dec 2014 Figure 8 provides the detail of cancellations by day from October 2013 to the end of December 2014. Peaks in December 2013 were as a result of the telephone failure in the London ACC followed by weather (mainly fog) in the following days. An airline strike at the start of April 2014 can be observed where the cancellation rate rose to 8%. In May French ATC industrial action generated small peaks at 2%. Looking further into detail Q4 2014 saw industrial action at Lufthansa on 1, 2 and 4 December and at TAP Air Portugal on 2 December resulted in a reduction of approximately 1,600 flights with the rate of cancellations reaching 10%. Industrial action also took place in Belgium, resulting in approximately 300 flights being cancelled on 8 December and 800 flights on 15 December. Industrial action occurred in Italy on the 12 December where 1,000 flights did not fly as consequence of the action. There was also a flight data server failure at London ACC on 12 December with London Heathrow being impacted although the effect was more of delays rather than cancellations. EUROCONTROL 5 CODA

3. CODA Scheduling Indicators Scheduling correctly is a difficult art: if too long a time is blocked for a flight, the airline will not be able to make best use of resources - staff, airframes, infrastructure. But too short a time can arguably be worse as late flights generate rotational delay with late incoming aircraft and passengers from previous flights having to be accommodated. When flights leave on time but arrive after the scheduled time of arrival they cause reactionary delays. Schedule padding is essential for air carriers in order to find schedules which work with the typical patterns of delay, so that they can deliver passengers on time, and get maximum use out of their aircraft. Consequently when delays decrease it takes one or two (IATA) seasons for the airline to adapt its schedule accordingly. Two CODA scheduling indicators help airline schedulers determine the optimal schedule based on historical flight data: The Block Time Overshoot (BTO) or the percentage of flights with an actual block time which exceeds the scheduled block time. The European BTO in Q4 2014 was 29%, an increase of 0.7 points when compared to Q4 2013, indicating that more flights had an actual block time exceeding their scheduled block time during the quarter, this is also highlighted in the arrival punctuality (Figure 18). The Delay Difference Indicator - Flight (DDI-F) or the difference between departure and arrival punctuality expressed in minutes. This can be indicated as a positive or negative figure, for example a flight departing with 20 minutes delay and arriving with 30 minutes arrival delay will have a DDI-F of +10 minutes. The European DDI-F in Q4 2014 was -3.6 minutes, this was a decrease of 0.2 minutes per flight when compared to the fourth quarter of 2014. Figure 9. Block Time Overshoot (BTO) and Delay Difference Indicator - Flight (DDI-F) Q4 2014 vs. Q4 2013 EUROCONTROL 6 CODA

4. CODA Reactionary Delay Analysis As seen in the headlines section of this report the average delay per flight increased by 3% to 9.2 minutes per flight when compared to Q4 2013, the share of reactionary delay slightly increased stable. Reactionary delays are generated as a result of an aircraft s late arrival from a previous flight which in turn affects the punctuality of its next flight with the same aircraft as well as potentially delaying connecting passengers. See IATA Standard Delay Codes section for an overview of the IATA reactionary delay codes. Subsequently there are two types of reactionary delays: firstly as a result of the same aircraft being delayed on its next flight (rotational delay) and secondly when another aircraft is delayed as a result of another aircraft typically through passengers, crew and load connection (non-rotational delays). Reactionary delays are particually important to an individual airline s operational performance as an aircraft operating behind schedule may affect other flights of the airline as well as the overall European network. This section concentrates on IATA delay code 93 rotational delays, as these have a significant share of overall reactionary delays (approximately 90%) and the largest effect on network performance and passenger experience. Using enriched CODA data on intra-european flights we are able to match the aircraft registration to callsigns and link individual aircraft rotations by day. To build the sequence the aircraft are grouped by their individual registrations, then by their actual reported off block times. Following that, the arrival airport from one flight and the departure airport of the next leg is also matched. The sequence is ended when the scheduled ground time exceeds a pre-set limit dependent on the aircraft seat capacity. In this section of the Digest we are analysing the depth of the scheduled flight sequence meaning the number of flights operated per sequence. Further to this analysis we also analyse the share of rotational delay minutes and their contribution to the average delay per flight for all-causes. In Figure 10 analysis shows that the median number of scheduled flights per sequence for an aircraft is 3. This is in line with the trend observed during last IATA winter season (November to March 13) where the median also fell to 3, as fewer flights may be scheduled during this period. The percentage share of rotational delay was 38% for the quarter following an increase in airline related delay particually in November. Figure 10. Median Scheduled Flight Sequence Depth and Share of Reactionary Departure Delay on Intra European Flights Oct 2014 December 2014 EUROCONTROL 7 CODA

5. Average Delay per Flight for Departures In Q4 2014 the average delay per flight from all causes of delay (calculated as the difference between STD and AOBT)* increased from 9.6 to 9.7 minutes per flight when compared to Q4 2013. The average delay per delayed flight (ADD) fell by 1.9 minutes to 25 minutes. 36% of flights were delayed on departure (>=5 minutes) this was an increase of 1.6 points in comparison to that observed in Q4 2013. Figure 11. Average delay per flight (All-Causes) for Departures Figure 12. Average delay per flight (All-Causes) for Departures (STD vs. AOBT) *Figure 12 Average delay per flight (All-Causes) for Departures (STD vs. AOBT) is calculated as the difference between STD and AOBT. Figure 3 (Delay Causes Q4 2014 vs. Q4 2013) graph is calculated using the sum of the provided delay times by the airline. Figure 13. Percentage of Delayed Flights (All-Causes) for Departures EUROCONTROL 8 CODA

6. NM versus Aircraft Operator experience of Delay Figure 14. Breakdown of Average Delay per Flight (Network Manager vs. Aircraft Operator) This section presents the air transport delay situation as seen from the airlines and passengers viewpoint (the difference between the scheduled time of departure and the actual time of departure from the gate). Data coverage is 64% of commercial flights in the ECAC region for Q4 2014. Further analysis of airline data shows that the average en-route ATFM delay from all-causes was 0.3 minutes per flight. This was the same as the NM recorded average en-route ATFM delay of 0.3 minutes per flight. ATFM delays reported by airlines could be lower than the NM-calculated ATFM delays due to difference in methods: ATFM delays of NM are the (flight) planned delays expressed as a restriction; the airlines report the actual experienced ATFM delay on departure. For instance, a flight with an ATFM restriction may also have a handling delay absorbed within the ATFM restriction. For the airline, a part of this delay is the ATFM delay and the rest is the handling delay. According to airline reports, primary delays (eg. passenger and ramp handling delays) counted for 56% (5.4 min/flt) of which (0.3 min/flt) was attributed to en-route ATFM restrictions, with reactionary delays representing the remaining share of 44% at (4.2 min/flt). EUROCONTROL 9 CODA

7. Average Delay per Flight by Hour Figure 15. Breakdown of the Average Delay per Flight by Hour of the Day Q4 2014 (top) vs. Q4 2013 (bottom) During the first rotation phase (between 0300 and 0900 UTC) Figure 15 shows a slightly higher share of airline related delay in the early morning with the share remaining just under 40% then falling to 35% until 1500 UTC. Weather delay share increased in comparison to Q4 2013 and in a similar trend to Q4 2013 towards the end of the operational day reactionary delay peaked at 62%. EUROCONTROL 10 CODA

8. Average Delay per Delayed Flight >5 Minutes Figure 16. ADD vs. PDF on Departure Figure 17. ADD vs. PDF on Arrival The average delay per delayed flight on departure (ADDD) was 25 minutes, this was a decrease of 1.9 minutes in comparison to Q4 2013. In Q4 2014, 37% of flights were delayed on departure (>=5 minutes) this was an increase of 1.6 percentage points in comparison to Q4 2013. The higher percentages of delayed flights observed in October and December can be noted in the above graph. Regarding arrivals, the average delay per delayed arrival in Q4 2014 was 27 minutes (28 minutes per flight in Q4 2013). The percentage of delayed flights on arrival increased to 34% compared to 33% in Q4 2013. EUROCONTROL 11 CODA

9. Distribution of All Flights by Length of Delay (Punctuality) In Q4 2014, 45% of flights departed within the 5 minute threshold before or after the scheduled departure time (STD) an increase of 0.1 percentage points when compared to Q4 2013. Flights delayed >30 minutes from allcauses increased, these by 0.1 percentage points to 7.4%, however flights delayed >60 minutes increased by 0.2 points. Figure 18. All-Causes Departure and Arrival Punctuality Q4 2014 vs. Q4 2013 In regard to arrival punctuality, 27% of flights arrived on time within the 5 minute threshold before or after the scheduled arrival time, with longer delays (those >30 minutes) increasing to 7.6% compared to 7.5% in Q4 2014. In a change to the trend higher levels observed throughout 2013, flights arriving >15 minutes ahead of schedule decreased from 11% to 9.5% following the slight increase in delay observed during the quarter. EUROCONTROL 12 CODA

10. Average Delay per Flight for Arrivals The average delay per flight on arrival from all causes was 9.2 minutes per flight in Q4 2014 with the percentage of delayed flights increasing by 0.9 percentage points to 33.9% in comparison to Q4 2013. In Figure 22 out of the top 50 airlines by the number of flights (64% coverage of IFR flights, long haul and short haul operations included). Following the increase in delays during the quarter, 21 airlines saw more than 85% of their flights arrive within 15 minutes of the scheduled time of arrival vs. 26 airlines in Q4 2013. Figure 19. Average delay per flight (All- Causes) for Arrivals Figure 20. Average delay per flight (All-Causes) for Arrivals Figure 21. Percentage of Delayed Flights (All-Causes) for Arrivals Figure 22: *Top 50 Airlines by Arrival Punctuality Q4 2014 *The top 50 airlines reporting to CODA by number of flights. Arrival punctuality is calculated as the share of flights arriving within 15 minutes of the scheduled time (time of arrival at gate) EUROCONTROL 13 CODA

11. Top 20 Affected Departure Airports Figure 23. All-Causes Delay. Top 20 Affected Departure Airports Q4 2014 Rank Departure Airport ICAO Code Average delay per departure (mins) Average Delay per Flight Percentage Change Average Delay per Delayed Departure Percentage of Delayed Departures 1 LISBOA LPPT 15.9 27% 29.1 54.5% 2 ROME FIUMICINO LIRF 13.4 21% 23.4 57.4% 3 LONDON/HEATHROW EGLL 13.3-3% 25.4 52.5% 4 LONDON/GATWICK EGKK 13.0 1% 26.6 49.0% 5 MANCHESTER EGCC 12.0 14% 27.5 43.5% 6 ZURICH LSZH 11.6 29% 21.3 54.5% 7 PARIS CH DE GAULLE LFPG 11.4 4% 24.2 46.8% 8 FRANKFURT MAIN EDDF 11.2 34% 21.5 52.0% 9 SCHIPHOL AMSTERDAM EHAM 11.2 8% 24.1 46.2% 10 LAS PALMAS GCLP 10.5 6% 27.9 37.6% 11 BIRMINGHAM EGBB 10.5 17% 25.3 41.4% 12 MILANO MALPENSA LIMC 10.4 7% 28.4 36.4% 13 BRUSSELS NATIONAL EBBR 10.2 21% 23.3 44.0% 14 KOELN-BONN EDDK 9.9 6% 24.9 39.7% 15 BERGEN/FLESLAND ENBR 9.8-10% 36.0 27.3% 16 GENEVE COINTRIN LSGG 9.8 1% 26.9 36.5% 17 TEGEL-BERLIN EDDT 9.5 10% 23.3 40.6% 18 MALAGA LEMG 9.4 2% 25.4 37.2% 19 EDINBURGH EGPH 9.3 16% 28.5 32.7% 20 NICE LFMN 9.1-3% 25.8 35.5% Analysis of the Top 20 Affected Airports on departure shows that Lisbon airport ranked the highest with an average delay per flight of 16 minutes in Q4 2014, with high shares of airline and reactionary delay that affected departing flights. Rome Fiumicino ranked second with 13.4 minutes per flight, with a high share of flights experiencing weather delays. London Heathrow saw weather delays throughout the quarter, mainly winds including one day of disruption in October due to the remnants of hurricane Gonzalo. Figure 24. Main delay causes at the top 10 affected departure airports EUROCONTROL 14 CODA

12. Top 20 Affected Arrival Airports Figure 25. All-Causes Delay. Top 20 Affected Arrival Airports Q4 2014 Rank Arrival Airport ICAO Code Average delay per Flight (mins) Average Delay per Flight Percentage Change Average Delay per delayed arrival Percentage of delayed arrivals 1 LISBOA LPPT 15.5 38% 32.2 48.2% 2 LONDON/HEATHROW EGLL 13.5-11% 30.4 44.4% 3 LONDON/GATWICK EGKK 13.2 0% 37.8 34.9% 4 LONDON/STANSTED EGSS 11.3 2% 30.9 36.6% 5 MANCHESTER EGCC 10.9 1% 32.2 33.9% 6 KOELN-BONN EDDK 10.1 11% 28.9 34.8% 7 EDINBURGH EGPH 9.9 25% 30.1 32.8% 8 LAS PALMAS GCLP 9.8 28% 24.3 40.5% 9 MILANO MALPENSA LIMC 9.8 11% 31.4 31.2% 10 DUBLIN EIDW 9.7-21% 27.1 35.6% 11 ZURICH LSZH 9.6 34% 24.3 39.7% 12 BRUSSELS NATIONAL EBBR 9.6 16% 26.2 36.7% 13 ISTANBUL-ATATURK LTBA 9.5-12% 25.5 37.4% 14 FRANKFURT MAIN EDDF 9.5 29% 27.9 33.9% 15 SCHIPHOL AMSTERDAM EHAM 9.4 6% 30.3 30.9% 16 OSLO/GARDERMOEN ENGM 9.3-7% 28.0 33.2% 17 PARIS CH DE GAULLE LFPG 9.3 11% 27.0 34.4% 18 GENEVE COINTRIN LSGG 9.3-10% 26.7 34.8% 19 TEGEL-BERLIN EDDT 9.0 11% 26.3 34.4% 20 ROME FIUMICINO LIRF 8.9 40% 26.4 33.8% The Top 20 Affected Airports on arrival shows that Lisbon ranked the highest with an average delay per delayed flight of 32 minutes, with reactionary delays affecting arriving flights following delays experienced earlier in the day. The effect of the seasonal weather at London Heathrow can also be seen in Figure 26 with this airport ranking second, although average delays did fall by 11% to 14 minutes in the quarter. Figure 26. Main delay causes at the top 10 affected arrival airports EUROCONTROL 15 CODA

13. Top 20 Most Delayed Airport Pairs Figure 27. All-Causes Delay Situation for the 20 Most Delayed Airport Pairs Q4 2014 Rank Departure Airport Arrival Airport Average Delay Per Departure Change since Previous Period Average Delay Per Delayed Departure Percentage Delayed Departures 1 ROME FIUMICINO CATANIA FONTANAROSSA 15.0 30% 25.8 58.1% 2 LONDON/HEATHROW ZURICH 13.5 10% 22.7 59.6% 3 LONDON/HEATHROW FRANKFURT MAIN 13.1 14% 22.9 57.1% 4 LONDON/HEATHROW DUBLIN 12.8-1% 26.5 48.5% 5 SCHIPHOL AMSTERDAM LONDON/HEATHROW 12.5-6% 24.1 52.0% 6 ROME FIUMICINO PALERMO PUNTA RAISI 12.0-3% 22.3 53.6% 7 LONDON/HEATHROW MUENCHEN 2 11.8-6% 21.7 54.6% 8 ZURICH LONDON/HEATHROW 11.8 7% 22.6 52.0% 9 CATANIA FONTANAROSSA ROME FIUMICINO 11.7 44% 25.9 45.1% 10 LONDON/HEATHROW EDINBURGH 11.0 4% 23.2 47.5% 11 LONDON/HEATHROW GENEVE COINTRIN 10.8-11% 21.3 50.6% 12 DUBLIN LONDON/HEATHROW 10.6-15% 25.7 41.5% 13 LONDON/HEATHROW MADRID BARAJAS 10.5-28% 20.2 52.0% 14 GENEVE COINTRIN LONDON/HEATHROW 10.2-24% 25.0 40.7% 15 FRANKFURT MAIN LONDON/HEATHROW 10.2 7% 21.7 46.8% 16 PARIS CH DE GAULLE FRANKFURT MAIN 9.9 111% 19.9 49.6% 17 PARIS CH DE GAULLE LONDON/HEATHROW 9.8-15% 21.5 45.9% 18 TEGEL-BERLIN FRANKFURT MAIN 9.6 55% 23.3 41.1% 19 EDINBURGH LONDON/HEATHROW 9.6 14% 31.9 29.9% 20 LISBOA MADRID BARAJAS 9.5 24% 25.7 36.8% Analysis of the Top 20 most delayed city pairs show that Rome Fiumicino to Catania ranked at number one with an average delay per flight of 15 minutes in Q4 2014, flights on this route experienced reactionary delays. The effect of the weather can also be observed at London Heathrow during all 3 months of the quarter with the airport featuring in 12 times in the Top 20. EUROCONTROL 16 CODA

14. Year-on-Year Trends in All-Causes Delay Indicators This section summarises the year-on-year trends in the main indicators of delay from all causes. A flight is considered delayed from 5 minutes. This is based on CODA data covering 64% of commercial flights in the ECAC region in the fourth quarter of 2014. Figure 28. Average all-causes delay/delayed flight (departures left, arrivals right) Figure 29. Percentage of flights delayed for all-causes delay (departures left, arrivals right) Figure 30. Percentage of flights delayed >15mins for all-causes (departures left, arrivals right) EUROCONTROL 17 CODA

A. Glossary of Terms and Abbreviations Delay Parameter Abbreviations PDF TDF TDM TTF ADMD ADMD ADDD ADDD Percentage of Delayed Flights Total Delayed Flights Total Delay in Minutes Total Flights Average Delay per Flight on Departure Average Delay per Flight on Arrival Average Delay per Delayed Flight on Departure Average Delay per Delayed Flight on Arrival Glossary of Terms ACC AEA AHM AIBT AOBT ATFCM ATFM ATS BTO CODA DDI-F ECAC Area Control Centre Association of European Airlines Airport Handling Manual Actual In Block Time Actual Off Block Time Air Traffic Flow and Capacity Management Air Traffic Flow Management (used by IATA in the Standard IATA Delay Codes) Air Traffic Services Block Time Overshoot Central Office for Delay Analysis Delay Difference Indicator - Flight European Civil Aviation Conference ESRA08 EUROCONTROL Statistical Reference Area 2008 IATA NMD NMOC STA STD International Air Transport Association Network Manager Directorate NM Operations Centre (Network Manager) Scheduled Time of Arrival Scheduled Time of Departure EUROCONTROL 18 CODA

B. CODA Delay Groupings Link between CODA Causes and IATA delay codes. Primary Delay Causes CODA CAUSE Airline Airport Description IATA Code Passenger and Baggage 11-19 Cargo and Mail 21-29 Aircraft and Ramp Handling 31-39 Technical and Aircraft Equipment 41-49 Damage to Aircraft & EDP/Automated Equipment Failure 51-58 Flight Operations and Crewing 61-69 Other Airline Related Causes Others ATFM due to Restriction at Destination Airport 83 Airport Facilities 87 Restrictions at Airport of Destination 88 Restrictions at Airport of Departure 89 ATFM due to ATC En-Route Demand / Capacity 81 En-Route ATFM due to ATC Staff / Equipment En-Route 82 Governmental Security and Immigration 85-86 Weather (other than ATFM) 71-79 Weather ATFM due to Weather at Destination 84 Miscellaneous Miscellaneous 98-99 Reactionary Late Arrival of Aircraft, Crew, Passengers or Load 91-96 EUROCONTROL 19 CODA

Correlation between IATA Codes and the NM Regulation Codes Note: updated version published in the ATFCM Users Manual 18.1.1 1 1 http://www.eurocontrol.int/sites/default/files/content/documents/nm/network-operations/handbook/atfcm-users-manual-current.pdf EUROCONTROL 20 CODA

Standard IATA Delay Codes (AHM730) Others 00-05 AIRLINE INTERNAL CODES 06 (OA) NO GATE/STAND AVAILABILITY DUE TO OWN AIRLINE ACTIVITY 09 (SG) SCHEDULED GROUND TIME LESS THAN DECLARED MINIMUM GROUND TIME Passenger and Baggage 11 (PD) LATE CHECK-IN, acceptance after deadline 12 (PL) LATE CHECK-IN, congestions in check-in area 13 (PE) CHECK-IN ERROR, passenger and baggage 14 (PO) OVERSALES, booking errors 15 (PH) BOARDING, discrepancies and paging, missing checked-in passenger 16 (PS) COMMERCIAL PUBLICITY/PASSENGER CONVENIENCE, VIP, press, ground meals and missing personal items 17 (PC) CATERING ORDER, late or incorrect order given to supplier 18 (PB) BAGGAGE PROCESSING, sorting etc. 19 (PW) REDUCED MOBILITY, boarding / deboarding of passengers with reduced mobility. 21 (CD) DOCUMENTATION, errors etc. 22 (CP) LATE POSITIONING 23 (CC) LATE ACCEPTANCE 24 (CI) INADEQUATE PACKING 25 (CO) OVERSALES, booking errors 26 (CU) LATE PREPARATION IN WAREHOUSE 27 (CE) DOCUMENTATION, PACKING etc (Mail Only) 28 (CL) LATE POSITIONING (Mail Only) 29 (CA) LATE ACCEPTANCE (Mail Only) Cargo and Mail Aircraft and Ramp Handling 31 (GD) AIRCRAFT DOCUMENTATION LATE/INACCURATE, weight and balance, general declaration, pax manifest, etc. 32 (GL) LOADING/UNLOADING, bulky, special load, cabin load, lack of loading staff 33 (GE) LOADING EQUIPMENT, lack of or breakdown, e.g. container pallet loader, lack of staff 34 (GS) SERVICING EQUIPMENT, lack of or breakdown, lack of staff, e.g. steps 35 (GC) AIRCRAFT CLEANING 36 (GF) FUELLING/DEFUELLING, fuel supplier 37 (GB) CATERING, late delivery or loading 38 (GU) ULD, lack of or serviceability 39 (GT) TECHNICAL EQUIPMENT, lack of or breakdown, lack of staff, e.g. pushback Technical and Aircraft Equipment 41 (TD) AIRCRAFT DEFECTS. 42 (TM) SCHEDULED MAINTENANCE, late release. 43 (TN) NON-SCHEDULED MAINTENANCE, special checks and/or additional works beyond normal maintenance schedule. 44 (TS) SPARES AND MAINTENANCE EQUIPMENT, lack of or breakdown. 45 (TA) AOG SPARES, to be carried to another station. 46 (TC) AIRCRAFT CHANGE, for technical reasons. 47 (TL) STAND-BY AIRCRAFT, lack of planned stand-by aircraft for technical reasons. 48 (TV) SCHEDULED CABIN CONFIGURATION/VERSION ADJUSTMENTS. Damage to Aircraft & EDP/Automated Equipment Failure 51 (DF) DAMAGE DURING FLIGHT OPERATIONS, bird or lightning strike, turbulence, heavy or overweight landing, collision during taxiing 52 (DG) DAMAGE DURING GROUND OPERATIONS, collisions (other than during taxiing), loading/off-loading damage, contamination, towing, extreme weather conditions 55 (ED) DEPARTURE CONTROL 56 (EC) CARGO PREPARATION/DOCUMENTATION 57 (EF) FLIGHT PLANS 58 (EO) OTHER AUTOMATED SYSTEM EUROCONTROL 21 CODA

Flight Operations and Crewing 61 (FP) FLIGHT PLAN, late completion or change of, flight documentation 62 (FF) OPERATIONAL REQUIREMENTS, fuel, load alteration 63 (FT) LATE CREW BOARDING OR DEPARTURE PROCEDURES, other than connection and standby (flight deck or entire crew) 64 (FS) FLIGHT DECK CREW SHORTAGE, sickness, awaiting standby, flight time limitations, crew meals, valid visa, health documents, etc. 65 (FR) FLIGHT DECK CREW SPECIAL REQUEST, not within operational requirements 66 (FL) LATE CABIN CREW BOARDING OR DEPARTURE PROCEDURES, other than connection and standby 67 (FC) CABIN CREW SHORTAGE, sickness, awaiting standby, flight time limitations, crew meals, valid visa, health documents, etc. 68 (FA) CABIN CREW ERROR OR SPECIAL REQUEST, not within operational requirements 69 (FB) CAPTAIN REQUEST FOR SECURITY CHECK, extraordinary Weather 71 (WO) DEPARTURE STATION 72 (WT) DESTINATION STATION 73 (WR) EN ROUTE OR ALTERNATE 75 (WI) DE-ICING OF AIRCRAFT, removal of ice and/or snow, frost prevention excluding unserviceability of equipment 76 (WS) REMOVAL OF SNOW, ICE, WATER AND SAND FROM AIRPORT 77 (WG) GROUND HANDLING IMPAIRED BY ADVERSE WEATHER CONDITIONS ATFM + AIRPORT + GOVERNMENTAL AUTHORITIES AIR TRAFFIC FLOW MANAGEMENT RESTRICTIONS 81 (AT) ATFM due to ATC EN-ROUTE DEMAND/CAPACITY, standard demand/capacity problems 82 (AX) ATFM due to ATC STAFF/EQUIPMENT EN-ROUTE, reduced capacity caused by industrial action or staff shortage, equipment failure, military exercise or extraordinary demand due to capacity reduction in neighbouring area 83 (AE) ATFM due to RESTRICTION AT DESTINATION AIRPORT, airport and/or runway closed due to obstruction, industrial action, staff shortage, political unrest, noise abatement, night curfew, special flights 84 (AW) ATFM due to WEATHER AT DESTINATION AIRPORT AND GOVERNMENTAL AUTHORITIES 85 (AS) MANDATORY SECURITY 86 (AG) IMMIGRATION, CUSTOMS, HEALTH 87 (AF) AIRPORT FACILITIES, parking stands, ramp congestion, lighting, buildings, gate limitations, etc. 88 (AD) RESTRICTIONS AT AIRPORT OF DESTINATION, airport and/or runway closed due to obstruction, industrial action, staff shortage, political unrest, noise abatement, night curfew, special flights 89 (AM) RESTRICTIONS AT AIRPORT OF DEPARTURE WITH OR WITHOUT ATFM RESTRICTIONS, including Air Traffic Services, start-up and pushback, airport and/or runway closed due to obstruction or weather 2, industrial action, staff shortage, political unrest, noise abatement, night curfew, special flights Reactionary 91 (RL) LOAD CONNECTION, awaiting load from another flight 92 (RT) THROUGH CHECK-IN ERROR, passenger and baggage 93 (RA) AIRCRAFT ROTATION, late arrival of aircraft from another flight or previous sector 94 (RS) CABIN CREW ROTATION, awaiting cabin crew from another flight 95 (RC) CREW ROTATION, awaiting crew from another flight (flight deck or entire crew) 96 (RO) OPERATIONS CONTROL, re-routing, diversion, consolidation, aircraft change for reasons other than technical Miscellaneous 97 (MI) INDUSTRIAL ACTION WITH OWN AIRLINE 98 (MO) INDUSTRIAL ACTION OUTSIDE OWN AIRLINE, excluding ATS 99 (MX) OTHER REASON, not matching any code above SOURCE: IATA Airport Handling Manual (730 & 731) 2 Restriction due to weather in case of ATFM regulation only, else refer to code 71 (WO) EUROCONTROL 22 CODA

Standard IATA Delay Sub-Codes (AHM731) 73 (WR) WEATHER: EN ROUTE OR ALTERNATE Z OUTSIDE AIRCRAFT LIMITS Y OUTSIDE CREW LIMITS X ETOPS 81 (AT) ATFM DUE TO ATC EN-ROUTE DEMAND/CAPACITY, standard demand/capacity problems Z ATC ROUTEINGS Y HIGH DEMAND OR CAPACITY 82 (AX) ATFM DUE TO ATC STAFF/EQUIPMENT EN-ROUTE, reduced capacity caused by industrial action or shortage or equipment failure, extraordinary demand due to capacity reduction in neighbouring area Z INDUSTRIAL ACTION Y EQUIPMENT FAILURE X STAFF SHORTAGE W MILITARY ACTIVITY V SPECIAL EVENT 83 (AE) ATFM DUE TO RESTRICTION AT DESTINATION AIRPORT, airport and/or runway closed due to obstruction, industrial action, staff shortage, political unrest, noise abatement, night curfew, special flights Z HIGH DEMAND / ATC CAPACITY Y INDUSTRIAL ACTION X EQUIPMENT FAILURE W STAFF SHORTAGE V ACCIDENT / INCIDENT U MILITARY ACTIVITY T SPECIAL EVENT S NOISE ABATEMENT/NIGHT CURFEW R OTHER 87 (AF) AIRPORT FACILITIES, parking stands, ramp congestion, lighting, buildings, gate limitations, etc. Z LACK OF PARKING STANDS Y RAMP CONGESTION X LIGHTING OR BUILDINGS W GATE LIMITATION / NO GATE AVAILABLE V BAGGAGE SORTING SYSTEM DOWN / SLOW U NO PUSH BACK CLEARANCE DUE TO INFRASTRUCTURE T JET BRIDGE INOPERATIVE S LACK OF CHECK IN COUNTERS R ELECTRICAL SYSTEM FAILURE P PASSENGER TRANSPORT SYSTEM FAILURE N PUBLIC ADDRESS/FLIGHT INFORMATION DISPLAY SYSTEM FAILURE M INSUFFICIENT FIRE COVER L GROUND COMMUNICATION SYSTEM FAILURE K NO PUSH BACK CLEARANCE DUE TO CONSTRUCTION J BREAKDOWN OF AIRPORT FUELLING SYSTEM H LATE OR LACK OF FOLLOW ME FOR PUSH-BACK G ANY OF THE ABOVE AT THE DESTINATION AIRPORT 89 (AM) RESTRICTIONS AT AIRPORT OF DEPARTURE WITH OR WITHOUT ATFM RESTRICTIONS, including Air Traffic Services, start-up and pushback, airport and/or runway closed due to obstruction or weather. Z ATC CAPACITY Y ATC INDUSTRIAL ACTION X ATC STAFFING W ATC EQUIPMENT V ATC ACCIDENT/INCIDENT U ATC DUE TO DE-ICING T ATC SPECIAL EVENT S ATC WEATHER R ATC RESTRICTIONS DUE TO CURFEW P ATC POLITICAL UNREST N ATC ENVIRONMENTAL M AIRPORT CLOSURE L RUNWAY CLOSURE K START-UP/PUSHBACK CLEARENCE DELAY (LOCAL ATC) J LOST FLIGHT PLAN BY ATC H CONSTRUCTION WORK/MAINTENENCE G OTHER 93 (RA) AIRCRAFT ROTATION, late arrival of aircraft from another flight Z LATE ARRIVAL DUE DEPARTURE DELAY AT PREVIOUS STATION Y LATE ARRIVAL DUE ENROUTE DELAY X LATE ARRIVAL DUE DELAY AFTER LANDING W LATE ARRIVAL DUE TO HIGH DEMAND FOR DESTINATION STATION V LATE ARRIVAL DUE TO WEATHER AT DESTINATION U LATE ARRIVAL DUE TO TECHNICAL REASONS EUROCONTROL 23 CODA

CODA Coverage Of Commercial Flights in the ECAC Region Q4 2014 EUROCONTROL 24 CODA