Performance Review Report - PRR 2014

EUROCONTROL

PRR 2014

Performance Review Report

An Assessment of Air Traffic Management in Europe during the Calendar Year 2014

Performance Review Commission I May 2015

Background

This report has been produced by the Performance Review Commission (PRC). The PRC was established by the Permanent Commission of EUROCONTROL in accordance with the ECAC Institutional Strategy 1997. One objective of this strategy is “to introduce a strong, transparent and independent performance review and target setting system to facilitate more effective management of the European ATM system, encourage mutual accountability for system performance…”

All PRC publications are available from the website: http://www.eurocontrol.int/prc

Notice

The PRC has made every effort to ensure that the information and analysis contained in this document are as accurate and complete as possible. Only information from quoted sources has been used and information relating to named parties has been checked with the parties concerned. Despite these precautions, should you find any errors or inconsistencies we would be grateful if you could please bring them to the PRU’s attention.

The PRU’s e-mail address is [email protected]

Copyright notice and Disclaimer

© European Organisation for the Safety of Air Navigation (EUROCONTROL)

This document is published by the Performance Review Commission in the interest of the exchange of information.

It may be copied in whole or in part providing that the copyright notice and disclaimer are included. The information contained in this document may not be modified without prior written permission from the Performance Review Commission.

The views expressed herein do not necessarily reflect the official views or policy of EUROCONTROL, which makes no warranty, either implied or express, for the information contained in this document, neither does it assume any legal liability or responsibility for the accuracy, completeness or usefulness of this information.

Printed by EUROCONTROL, 96, rue de la Fusée, B-1130 Brussels, Belgium. The PRC’s website address is http://www.eurocontrol.int/prc. The PRU’s e-mail address is [email protected].

EUROCONTROL

DOCUMENT IDENTIFICATION SHEET

DOCUMENT DESCRIPTION

Document Title

Performance Review Commission

Performance Review Report covering the calendar year 2014 (PRR 2014)

PROGRAMME REFERENCE INDEX: EDITION: EDITION DATE:

PRC Performance Review Report Final report 21‐May‐2015

SUMMARY

This report of the Performance Review Commission analyses the performance of the European Air Traffic Management System in 2014 under the Key Performance Areas of Safety, Capacity, Environment and Cost‐efficiency.

Keywords

Air Traffic Management Performance Measurement

Performance Indicators ATM ANS

CONTACT:

Performance Review Unit, EUROCONTROL, 96 Rue de la Fusée,

B‐1130 Brussels, Belgium. Tel: +32 2 729 3956, E‐Mail: [email protected]

http://www.eurocontrol.int/prc

DOCUMENT STATUS AND TYPE

STATUS DISTRIBUTION

Draft General Public

Proposed Issue EUROCONTROL Organisation

Released Issue Restricted

INTERNAL REFERENCE NAME: PRR 2014

EXECUTIVE SUMMARY

EXECUTIVE SUMMARY PRR 2014

i

EXECUTIVE SUMMARY

European ATM Performance

Key Performance Indicator Data & commentary

TRAFFIC

IFR flights Eurocontrol Variation

2014 9.60 M + 1.7%

After the decrease between 2011 and 2013, average daily IFR flights in Europe increased by +1.7% in 2014 with notable regional variations.

For 2015, the STATFOR 7‐year forecast (2015) expects European flights to grow by +1.5% in the baseline scenario.

SAFETY

Accidents with ANS contribution

Eurocontrol Variation

2014 0 ‐1

Overall, safety levels in Europe continue to remain high, yet with scope for further improvements in some areas.

Over the past five years, there were only two accidents where ANS was a factor in the causal chain of events.

CAPACITY

En‐route ATFM delay per flight


2014 0.61 +0.08

After the best year on record in 2013, en‐route ATFM delays increased again to 0.61 minutes per flight in 2014.

The most constraining ACCs in 2014 were Nicosia, Warsaw, Lisbon, Athinai/Macedonia, Canarias, Brest, Reims, and Marseille.

ENVIRONMEN

T

En‐route flight efficiency (vs. flight plan)


2014 4.70% ‐0.16%pt.

En‐route flight efficiency continued to improve in 2014. The level of inefficiency in filed flight plans in Europe decreased from 4.86% to 4.70% and actual trajectories improved at an even higher rate from 3.14% to 2.72%.

COST‐EFFICIENCY

En‐route ANS costs per SU (€2009)


2013 53.2 ‐3.3%

In 2013, en‐route ANS costs decreased by 1.3% while en‐route service units grew by +2.1% leading to an overall decrease of en‐route unit costs by ‐3.3% compared to 2012.

‐6%

‐4%

‐2%

0%

2%

4%

6%

8%

10%

5

6

7

8

9

10

11

12

13

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

2020

% annual growth (bars)

IFR flights (million)

STATFOR (Feb. 2015)( before 1997, estimation based on Euro 88 traffic variation)

Source : EUROCONTROL/STATFOR (ESRA2008)

Feb. 2008

Feb. 2011forecast

4 3 3 1 2 1 2 1 10

10

20

30

40

50

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Number of accidents

Source: EASA Accidents with ANS contribution

Total commercial air transport (CAT) accidents and accidents with ANS

2.2

2.9

4.5

2.9

2.5

1.4

0.9

0.8 0.9 1.0 1.2 1.4

0.9

2.0

1.1

0.63

0.53

0.61

70

80

90

100

110

120

130

140

150

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Traffic index (base: 1997)

En‐route ATFM delay/ flight (m

in.)

ATC Capacity & Staffing ATC Other (strike, equipment, etc.)WEATHER OTHER (Special event, military, etc.)IFR Traffic

Average en route ATFM delay per flight (EUROCONTROL area)

Source: Network Manager

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

JAN

FEB

MAR

APR

MAY

JUN

JUL

AUG

SEP

OCT

NOV

DEC

JAN

FEB

MAR

APR

MAY

JUN

JUL

AUG

SEP

OCT

NOV

DEC

JAN

FEB

MAR

APR

MAY

JUN

JUL

AUG

SEP

OCT

NOV

DEC

2012 2013 2014

Flight plan Actual trajectory

4.864.70

3.14

2.72

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

2011

2012

2013

2014

Source: PRU analysis

inef

ficie

ncy

(%)

60.1 56.7

53.8 55.0 53.2 53.2

97 97 98 97

100

103

108 107

109 113

20

30

40

50

60

2009Actuals

2010Actuals

2011Actuals

2012Actuals

2013Actuals

2014Forecasts

En‐route real cost

per SU (€2009)Real en‐route

unit costs per SU for the

EUROCONTROL Area (€2009)

En route SU index (2009)

En route ANS cost index (2009)


EXECUTIVE SUMMARY


ii

Introduction PRR 2014 presents an assessment of the performance of European Air Navigation Services (ANS) for the calendar year 2014.

ANS in European Air Transport After the decrease between 2011 and 2013, flights in Europe increased again by +1.7% in 2014 with a positive medium term outlook. According to the latest STATFOR 7‐year forecast (2015), flights are expected to grow by 1.5% in 2015 and to continue with an annual average growth rate of 2.5% between 2014 and 2021.

However, despite the positive growth in 2014 and the promising outlook, the impact of the economic crisis on the industry is still prominent. At European level, there were almost three million flights less in 2014 than initially predicted before the economic crisis in 2008.

Whereas the number of flights still remains below 2008 levels in 2014, there has been a continuous increase in average aircraft size and passenger numbers during the period which suggests that airlines responded to the crisis with a reduction in the number of services but with, on average, larger aircraft. Hence, as a result of the increase in average aircraft weight and longer flights, en‐route service units (+5.9%) grew notably stronger than flights (+1.7%) in 2014.

With the exception of Albania, Armenia, Malta, Moldova and Ukraine, all EUROCONTROL states showed a positive growth in 2014, with Turkey remaining the main contributor to the growth in Europe. Four of the five largest States (Germany, Italy, UK, and Spain) experienced a traffic growth close to the high forecast scenario in 2014 and traffic was even higher than predicted in the high forecast scenario in Greece and Cyprus.

European traffic was affected by a number of foreseen and unforeseen events in 2014. The tragic loss of MH17, in Ukrainian airspace in July was undoubtedly the most significant event in European airspace during 2014. As a result, there was a notable shift of traffic. While traffic in Ukraine and Moldova dropped by ‐36.9% and ‐24.2% respectively, traffic in Bulgaria (24.1%), Romania (16.6%), Hungary (11.6%), Turkey (11.2%), and Slovakia (9.8%) increased far beyond forecast levels but could be accommodated without noteworthy delays.

Worldwide, airline safety improved to the best level on record with a rate of one fatal accident per 2.38 million flights in 2014. In the EUROCONTROL area safety levels remain high although preliminary 2014 data suggests a slight increase in total commercial air transport accidents and ANS‐related incidents in 2014 following the lowest level on record in 2013.

Compared to 2013 (which was the best year on record), arrival punctuality in Europe decreased slightly from 84.0% to 83.7% in 2014. Reactionary delay, caused by delay which could not be absorbed on subsequent flight legs, remains the largest single delay group (44.5%) in 2014, followed by delays due to turn round issues (37%). The ANS‐related share (en‐route and airport combined) in total departure delays was 13.3% in 2014.

Although the decrease in arrival punctuality in 2014 was mainly driven by a deterioration in turn‐round performance at airports, en‐route ATFM delays also increased compared to 2013. Due to this increase in en‐route ATFM delay in 2014 and an increase in flights at slightly better efficiency levels (taxi‐out, en‐route, terminal) than in 2013, estimated total additional ANS‐related operating time increased by 2.4% in 2014.

The total economic evaluation of ANS performance presents a consolidated view of direct ANS costs and estimated indirect ANS‐related costs (ATFM delays, additional taxi‐out and ASMA time, horizontal en route flight efficiency) borne by airspace users.

In the Single European Sky (SES) area, actual en‐route and terminal ANS costs in 2013 were lower than initially projected in last year’s PRR, suggesting that ANSPs were able to adjust their costs to the ‐1.3% decline in traffic in 2013. Hence, combined with the notable lower delay related costs observed in 2013, total economic costs decreased by ‐3.5% compared to 2012 (initial projection for 2013 was +0.9%).

Based on the latest available ANS cost projections for 2014, estimated total economic ANS‐related costs in the SES area are estimated to increase by 2.4% compared to 2013. The increase is mainly driven by the projected increase in en‐route and terminal ANS costs and the notable increase in en‐route ATFM delays in 2014.

EXECUTIVE SUMMARY


iii

Safety (2013/2014) Overall, safety levels in Europe continue to remain high, yet with scope for further improvements in some areas.

Over the past four years (2011‐2014) there were no fatal ANS‐related accidents and only one accident with ANS contribution (MET related) in 2013. The number of serious incidents with ANS contribution continued the positive trend observed since 2010 and decreased further in 2014 to the lowest level on record. Separation minima infringements remained the single largest category for serious ANS‐related incidents, followed by runway incursions. Although considerable progress has been made over the past years, there is still a need to further improve safety data quality and completeness.

After the continuous increase in reporting levels between 2004 and 2012, the total number of incidents reported fell again by 17% in 2013. However, with the available data it is difficult to differentiate whether the decrease in 2013 is due to genuine safety performance improvements or a drop in the level of reporting. Overall, final 2013 data was received from 36 the reporting from EUROCONTROL Member States via the Annual Summary Template (AST) reporting mechanism and the preliminary 2014 data was submitted by a record number of 40 States.

Although already flagged up as a crucial area for improvement in previous PRRs, the share of unclassified incidents, as reported through the AST mechanism, remained unacceptably high in 2013 even though improvements are visible. There is an urgent need for further action in order to closely monitor trends and to support the prevention of similar incidents. Lastly, more effort should be put in place by the Member States to ensure that the number of ATM related occurrences not severity classified, continues to decrease.

The need to accelerate the deployment of automatic safety data monitoring to complement manual reporting was also already addressed in last year’s report. The latest available information from the Network Manager suggests that 14 Member States now use ASMT which represents an increase of two States compared to 2013. However, there is a concern that the implementation of A‐SMGCS Level 1 (improved surveillance) lags behind. By 2014, only 52% of the 46 airports have achieved full operational capability. Being a prerequisite for the implementation of A‐SMGCS Level 2 (Surveillance + Safety Nets), this indicates that additional work is still needed in order to achieve the objectives set in the ATM Master Plan by the end of 2017 and avoid delayed implementation of Level 2. The use of A‐SMGCS or other equivalent runway safety programmes to improve reporting and safety performance is strongly encouraged by the PRC.

To ensure that safety remains an integral part in the evolution of the more and more complex aviation system, States and service providers are encouraged to proactively engage in safety risk modelling. The improved understanding of how the various elements of the ATM system contribute to overall safety levels not only helps to better identify risk areas today but also how modifications to the system could improve performance in the future.

Operational En-route ANS Performance (2014) After the continuous improvement between 2010 and 2013, (reaching the lowest level of en‐route ATFM delay per flight on record in 2013 (0.53 minutes per flight)), en‐route ATFM delays in the EUROCONTROL area increased again to 0.61 minutes per flight in 2014.

The performance deterioration in 2014 was mainly attributed to ATC capacity issues confirming concerns already highlighted in PRR 2013 that ATFM delays could spike, when traffic grows again, unless sufficient attention is focussed on capacity planning and deployment.

In 2014, European traffic flows and capacity management were affected by a number of significant events including the downing of Malaysian Airline MH17 in Ukrainian airspace in July. The ensuing total closure of airspace in the eastern part of Ukraine generated many re‐routings in that region. Traffic in Bulgaria, Romania, Hungary, Turkey, and Slovakia increased far beyond forecast levels but, thanks to the efforts of those ANSPs, they were accommodated without significant delays.

While capacity constraints can occur from time to time, area control centres (ACCs) should not generate high delays on a regular basis. In 2014, the most constraining ACCs were Nicosia (failure to implement capacity plans), Warsaw (issues with the implementation of the new PEGASUS 21 system), Lisbon (reoccurring issues in November), Canarias (capacity planning and delay classification issues), Athinai/Macedonia (insufficient capacity

EXECUTIVE SUMMARY


iv

in summer), Reims (capacity planning), Brest and Marseille (ATC industrial action) which accounted together for more than half of all European en‐route ATFM delays (54.6% ) but only for 17.8% of total flight hours controlled.

It is essential, particularly in view of the considerable lead times, to carefully plan and also deploy capacity in line with projected traffic growth. Over‐conservative capacity planning removes buffers against unexpected traffic increases and increases the risk of significant disruption to aircraft operations.

Although the unexpected changes in traffic flows in 2014 had a moderate negative effect on European flight efficiency in 2014, overall, there was a notable improvement in 2014, confirming the positive trend observed over the past years. Coordinated by the Network Manager, 30 of the 64 ACCs had implemented various steps of Free Route Operations by the end of 2014 with clear benefits in terms of flight efficiency.

In 2014, the level of inefficiency in filed flight plans in Europe decreased from 4.86% to 4.70% and actual trajectories improved at an even higher rate from 3.14% to 2.72%. Whereas the stronger improvement in actual trajectories is positive in terms of fuel burn and CO2 emissions, the widening of the gap between planned and actual flown trajectories needs to be monitored closely as it has a negative impact on network predictability.

A voluntary review of only three Member States identified significant differences, not only between the national arrangements and the FUA specification but also between the individual Member States, in how the airspace is managed to provide the optimum benefit for civil and military airspace users. A wider review could help to better identify and understand existing shortcomings and identify best practice for the future benefit of all airspace users.

Operational ANS Performance at Airports (2014) In 2014, IFR movements at the top 30 airports increased by +2.7% compared to the previous year, which is stronger than the European traffic growth in 2014 (+1.7% vs. 2013). Despite the growth in 2014, traffic at the top 30 airports remains 3.3% below 2008 levels. Compared to 2013, the number of passengers at the top 30 airports increased by +5.8% in 2014 (9.9% above 2008 levels).

There were notable differences in growth between airports in 2014. The Turkish airports continued their remarkable growth already observed in previous years and, following a traffic decrease at Paris Charles de Gaulle and Frankfurt airport, London Heathrow now ranks first in terms of average daily movements in Europe.

Operational cancellations at airports were included for the first time in this year’s report. Overall, around 1.5% of the flights confirmed by the air carrier the day before operations were cancelled in 2014. Data on cancellations were not available from all airports and thus limit the scope of the analysis of this report. With the on‐going establishment of the airport operator data flow specification, the level of reporting will improve in the future.

In order to avoid excessive disruptions in daily operations, demand at larger airports is already regulated strategically in terms of volume and concentration through the airport capacity declaration process. On the day of operations, ANS plays a key role in balancing traffic with available capacity at airports within the given infrastructural and environmental constraints and in the integration of airports in the European network. In view of the considerable downward adjustment of airport expansion plans in Europe, the optimised use of available capacity is crucial to keep delays to a minimum.

The indicators used for the evaluation of ANS‐related performance relate to the management of the inbound (i.e. arrival ATFM delay and ASMA additional time) and outbound (i.e. ATC pre‐departure delay and additional taxi out time) traffic flow and are consistent with the SES performance scheme.

In 2014, the average additional time for ASMA and taxi‐out improved slightly at the top 30 European airports whereas airport ATFM arrival delays and ATC pre‐departure delays increased at the same time.

London Heathrow, Frankfurt, Amsterdam Schiphol, Rome Fiumicino, Zurich, and the two Istanbul airports had the most notable impact on the network in 2014. The underlying reasons differ by airport and range from the airport scheduling process to the ability to sustain declared capacity in various circumstances on the day of operations.

Improved data exchange fostered by A‐CDM implementation and increased situation awareness are key enabler for performance improvements at airports. However the mere availability of information does not automatically resolve issues without proactively addressing relevant optimisation processes.

EXECUTIVE SUMMARY


v

For example, initial work on XMAN demonstrated positive improvements in term of reducing the additional ASMA time for London Heathrow. The better use of cross‐border arrival management (linking the terminal approach and en‐route function), is a key enabler for the absorption of possible holdings in a more environmental friendly manner, as the absorption takes place in the en‐route phase through reduced cruise speed.

On the other hand, local constraints at Rome Fiumicino pose a challenge to take advantage of the benefits of A‐CDM with a view to the optimisation of departure sequencing process.

The evaluation of daily taxi‐out performance shows that the current performance framework does not sufficiently address the impact of adverse weather on the performance indicators. Future work is required in order to better evaluate the level of efficiency in such conditions (e.g. de‐icing conditions).

ANS Cost-efficiency (2013) PRR2014 analyses performance in 2014 for all KPAs, except for cost‐efficiency, which analyses performance in 2013 as this is the latest year for which actual financial data are available.

The pan‐European system en route cost‐efficiency performance improved in 2013. Real en‐route costs per service unit decreased by ‐3.3% compared to 2012, reflecting a decrease of ‐1.3% in total costs and an increase of +2.1% in SUs. Total en‐route costs were also ‐5.6% lower than planned, responding to SUs which were ‐5.3% lower than forecast.

However, costs can be significantly impacted year on year by changes in (accounting) provisions for future liabilities (mainly for pensions). As highlighted in PRR 2013, the volatility of these (accounting) provisions raises concerns in the context of charging and performance, as changes in these provisions, which can be significant, do not necessarily represent costs directly attributable to the provision of ANS in the year in which they are recorded.

For the SES States, 2013 is the second year of application of the “determined costs” method with specific risk‐sharing arrangements. For the other nine EUROCONTROL States participating in the Route Charges System, the “full cost‐recovery method” continued to apply in 2013. The evaluation between SES states and the 9 other states does not yet show clear trends.

Under the determined costs method, the amounts ultimately paid by the airspace users differ from the actual costs due to the traffic risk‐sharing, cost‐sharing and other adjustments provided in the Charging Regulation. The PRC computes that, in 2013, these “true costs for users” are +3.5% higher than the actual costs of States/ANSPs but ‐2.0% lower than the determined costs provided for 2013 in the RP1 performance plans. This should be seen in the context of actual costs in 2013 which are ‐5.9% lower than planned in the RP1 performance plans. The fact that several States/ANSPs demonstrated lower costs than planned was duly considered during the RP2 assessment process in order to ensure that users benefit from the cost reductions achieved during RP1.

The outlook for 2014‐2019 suggests that the recent improved cost performance will not continue. Instead current plans/forecasts show costs stabilizing over the period. The forecast decrease in en‐route unit cost (‐2.2% per year on average between 2013 and 2019) is due to the forecast increase in SUs.

Terminal ANS unit costs (TNSUs) decreased by ‐3.6% between 2012 and 2013 reflecting a reduction of ‐3.9% in terminal ANS costs in real terms and a small traffic decrease (‐0.3%, TNSU).

The outlook for 2015‐19 shows a slower rate of decline in costs. Terminal ANS unit costs are forecast to decrease by ‐2.4% per year on average, reflecting an average reduction of ‐0.5% p.a. in costs and an average increase of +2.0% p.a. in TNSUs

Gate‐to‐gate unit economic costs (i.e. en‐route plus terminal ATM/CNS provision costs plus taking account of the cost of delay) decreased by ‐3.6% in 2013, reflecting both the reduction in ATFM delays unit costs compared to 2012 (‐18.2%) and a decrease in gate‐to‐gate unit ATM/CNS provision costs (‐1.9%). However the increased levels of ATFM delays observed in 2014 suggests that unit economic costs could be higher in 2014.

The benchmarking analysis of gate‐to‐gate unit ATM/CNS provision costs indicates that the reduction of ‐1.9% in 2013 compared to 2012 is mainly due to the reduction of ‐3.0% in support costs (70% of total costs). Over the 2009‐2013 period, pan‐European unit ATM/CNS provision costs significantly reduced by ‐2.2% p.a., reflecting a gradual decrease in unit support costs and increasing ATCO productivity while ATCO employment costs remained fairly constant.

EXECUTIVE SUMMARY


vi

PRC Recommendations 2014

Rationale for the recommendation

a. The PC is invited to note the PRC’s Performance Review Report for 2014 (PRR 2014) and to submit it to the Permanent Commission.

b. The PC is invited to request the Director General EUROCONTROL to investigate the factors contributing to the high number of poorly coded, unclassified and undetermined safety occurrences and to propose lines of action to PC 44 (December 2015) on how to improve the situation.

Although already flagged up as a crucial area for improvement in previous PRRs, the share of unclassified incidents, as reported through the AST mechanism, remained unacceptably high even though improvements are visible.

c. The PC is invited:

i. to request Member States, to task their ANSPs to develop and implement capacity plans which are, at a minimum, in line with the Reference Capacity Profile (from the NOP); and to ensure that capacity is made available during peak demand,

ii. to ask the Director General EUROCONTROL to report on those States that have insufficient capacity plans compared to the Reference Capacity Profile to PC 44 (December 2015).

Sufficient capacity plans must be developed and implemented in advance of traffic demand. It is evident that several Member States have downgraded and postponed existing capacity plans despite the predicted rise in future traffic figures. This short‐sighted behaviour promises serious disruption to the network and to aircraft operations, especially during peak traffic periods.

d. The PC is invited:

i. to request the PRC, in accordance with Article 10h of the PRC’s Terms of Reference, to review arrangements for civil military coordination and cooperation in the Member States by the end of 2015;

ii. to request the civil and military authorities in the Member States to assist the PRC to conduct this review;

iii. to ask the PRC to report to PC 44 (December 2015).

Review of only three Member States identified significant differences, not only between the national arrangements and the FUA specification but also between the individual Member States, in how the airspace is managed to provide the optimum benefit for civil and military airspace users.

vii

T A B L E O F C O N T E N T S

EXECUTIVE SUMMARY ..................................................................................................................... I

INTRODUCTION ........................................................................................................................................ II

ANS IN EUROPEAN AIR TRANSPORT ............................................................................................................. II

SAFETY (2013/2014) .............................................................................................................................. III

OPERATIONAL EN‐ROUTE ANS PERFORMANCE (2014) .................................................................................. III

OPERATIONAL ANS PERFORMANCE AT AIRPORTS (2014) ............................................................................... IV

ANS COST‐EFFICIENCY (2013) ................................................................................................................... V

PRC RECOMMENDATIONS 2014 ................................................................................................................ VI

PART I‐ BACKGROUND .................................................................................................................... 1

1 INTRODUCTION ...................................................................................................................... 1

1.1 PURPOSE OF THE REPORT ...............................................................................................................1

1.2 KEY EVENTS IN 2014 ....................................................................................................................1

1.3 CONSULTATION WITH STAKEHOLDERS ON PRR 2014 ..........................................................................2

1.4 STRUCTURE OF THE REPORT............................................................................................................2

1.5 REPORT SCOPE .............................................................................................................................2

1.6 IMPLEMENTATION STATUS OF PRC RECOMMENDATIONS .....................................................................3

1.7 PRC AS PERFORMANCE REVIEW BODY OF THE SINGLE EUROPEAN SKY ...................................................3

2 ANS IN EUROPEAN AIR TRANSPORT ........................................................................................ 4

2.1 INTRODUCTION ............................................................................................................................4

2.2 EUROPEAN AIR TRAFFIC DEMAND ...................................................................................................4

2.3 SAFETY .....................................................................................................................................11

2.4 SERVICE QUALITY ........................................................................................................................12

2.5 ECONOMIC EVALUATION OF ANS PERFORMANCE .............................................................................17

2.6 CONCLUSIONS ...........................................................................................................................19

PART II – KEY PERFORMANCE AREAS ............................................................................................ 21

3 SAFETY ................................................................................................................................. 21

3.1 INTRODUCTION ..........................................................................................................................21

3.2 ACCIDENTS AND SERIOUS INCIDENTS (ANS‐RELATED AND WITH ANS CONTRIBUTION) ............................22

3.3 ATM‐RELATED INCIDENTS ...........................................................................................................24

3.4 REPORTING AND INVESTIGATION ...................................................................................................26

3.5 AVIATION SAFETY RISK ................................................................................................................30

3.6 CONCLUSIONS ...........................................................................................................................31

4 OPERATIONAL EN‐ROUTE ANS PERFORMANCE ..................................................................... 32

4.1 INTRODUCTION ..........................................................................................................................32

4.2 EN‐ROUTE ATFM DELAYS ............................................................................................................32

viii

4.3 EN‐ROUTE FLIGHT EFFICIENCY ......................................................................................................41

4.4 THE EFFECT OF THE UKRAINIAN CRISIS ON NETWORK CAPACITY & NETWORK FLIGHT EFFICIENCY ...............46

4.5 FLEXIBLE USE OF AIRSPACE ...........................................................................................................47

4.6 EUROPEAN ATFM PERFORMANCE (NETWORK LEVEL) .......................................................................50

4.7 CONCLUSIONS ...........................................................................................................................51

5 OPERATIONAL ANS PERFORMANCE AT AIRPORTS ................................................................. 52

5.1 INTRODUCTION ..........................................................................................................................52

5.2 TRAFFIC EVOLUTION AT THE MAIN EUROPEAN AIRPORTS ....................................................................53

5.3 CANCELLATIONS AT EUROPEAN AIRPORTS .......................................................................................54

5.4 BALANCING CAPACITY WITH DEMAND AT AIRPORTS ..........................................................................55

5.5 MANAGEMENT OF THE ARRIVAL FLOW ...........................................................................................57

5.6 MANAGEMENT OF THE DEPARTURE FLOW .......................................................................................63

5.7 CONCLUSIONS ...........................................................................................................................67

6 ANS COST‐EFFICIENCY ........................................................................................................... 68

6.1 INTRODUCTION ..........................................................................................................................68

6.2 EN‐ROUTE ANS COST‐EFFICIENCY PERFORMANCE ............................................................................69

6.3 TERMINAL ANS COST‐EFFICIENCY PERFORMANCE .............................................................................77

6.4 ANSPS GATE‐TO‐GATE ECONOMIC PERFORMANCE ...........................................................................82

6.5 CONCLUSIONS ...........................................................................................................................87

ANNEX I ‐ IMPLEMENTATION OF PC DECISIONS ............................................................................. 88

ANNEX II ‐ ACC TRAFFIC AND DELAY DATA (2012‐2014) ................................................................. 91

ANNEX III ‐ TRAFFIC COMPLEXITY SCORES IN 2014 ........................................................................ 92

ANNEX IV ‐ FRAMEWORK : SERVICE QUALITY ................................................................................ 94

ANNEX V ‐ FRAMEWORK: ECONOMIC EVALUATION OF ANS PERFORMANCE ................................. 96

ANNEX VI ‐ AVIATION RISK MODEL ............................................................................................... 99

ANNEX VII ‐ AIRPORT TRAFFIC AND AIRPORT ANS PERFORMANCE RECORDS .............................. 100

ANNEX VIII ‐ RECONCILIATION WITH PRB 2013 MONITORING REPORT ........................................ 101

ANNEX IX ‐ GLOSSARY ................................................................................................................. 102

ANNEX X ‐ REFERENCES .............................................................................................................. 107

ix

L I S T O F F I G U R E S

Figure 1‐1: EUROCONTROL States (2014) ................................................................................................2

Figure 2‐1: Evolution of European IFR flights (1990‐2021) ......................................................................5

Figure 2‐2: European traffic indices (2004‐14) ........................................................................................5

Figure 2‐3: IFR flights by market segment (2004‐2014) ...........................................................................6

Figure 2‐4: Traffic variation by State (2014/2013) ...................................................................................6

Figure 2‐5: Actual IFR traffic growth vs forecast growth by State ...........................................................7

Figure 2‐6: Forecast traffic growth 2014‐2021 ........................................................................................8

Figure 2‐7: Peak day vs. avg. day traffic (Europe) ....................................................................................8

Figure 2‐8: Seasonal traffic variations at ATC‐Unit level (2014) ..............................................................9

Figure 2‐9: Complexity trend over time (Europe) ..................................................................................10

Figure 2‐10: Structural complexity and adjusted density at ANSP level (2014) ....................................10

Figure 2‐11: Accidents in EUROCONTROL area with ANS contribution (2003‐14) ................................11

Figure 2‐12: Serious Incidents in EUROCONTROL area with ANS contribution (2003‐14) .....................11

Figure 2‐13: European On time performance (2004‐14) .......................................................................12

Figure 2‐14: Variability of flight phases (2008‐14) .................................................................................13

Figure 2‐15: Evolution of delays and block times (2004‐14) ..................................................................13

Figure 2‐16: Departure delays by cause (2008‐2014) ............................................................................14

Figure 2‐17: Sensitivity to primary delays ..............................................................................................14

Figure 2‐18: Estimated ANS‐related impact on fuel burn/environment ...............................................16

Figure 2‐19: Share of ANS charges in airline operating costs (2013) .....................................................17

Figure 2‐20: Long‐term trend of traffic, unit costs and en‐route ATFM delay (summer) ......................17

Figure 2‐21: Estimated total economic ANS‐related costs (SES States) .................................................18

Figure 3‐1: Sources and scope of ANS safety review in this chapter .....................................................21

Figure 3‐2: Accidents (ANS‐related and with ANS contribution) in the EUROCONTROL area ...............22

Figure 3‐3: ANS‐related accidents by occurrence category (EUROCONTROL area) ...............................22

Figure 3‐4: Serious incidents (ANS‐related and with ANS contribution) in the EUROCONTROL area ...23

Figure 3‐5: ANS‐related serious incidents by occurrence category (EUROCONTROL area) ...................23

Figure 3‐6: Reported high‐risk SMIs in EUROCONTROL States (2004‐14P) ...........................................24

Figure 3‐7: Reported high‐risk UPAs in EUROCONTROL States (2004‐2014P) .......................................25

Figure 3‐8: Reported high‐risk RIs in EUROCONTROL States (2004‐2014P) ..........................................25

Figure 3‐9: Reported high‐risk ATM Spec. Occurrences in EUROCONTROL States (2004‐2014P) .........26

Figure 3‐10: Total number of reports and level of reporting (2004‐2014P) ..........................................27

Figure 3‐11: Severity not classified or not determined (2005‐2014P) ...................................................27

Figure 3‐12: Completeness of AST reported data in 2014 .....................................................................28

Figure 3‐13: States with automatic reporting tools (2014) ....................................................................29

Figure 3‐14: A‐SMGCS implementation status ......................................................................................29

x

Figure 3‐15: Airports late for A‐SMGCS Level 1 implementation ..........................................................30

Figure 4‐1: Average en‐route ATFM delay (1997‐2014) ........................................................................32

Figure 4‐2: En‐route delay per flight by classification ............................................................................33

Figure 4‐3: Monthly evolution of en‐route ATFM delays (2010‐2014) ..................................................33

Figure 4‐4: Overview of most constraining ACCs ...................................................................................34

Figure 4‐5: Delay categories for most constraining ACCs ......................................................................34

Figure 4‐6: Monthly ATFM en‐route delay 2014 and capacity planning (Nicosia) .................................35

Figure 4‐7: Monthly ATFM en‐route delay and hourly traffic profile (Warsaw) ....................................35

Figure 4‐8: Monthly ATFM en‐route delay (Lisbon) ...............................................................................36

Figure 4‐9: Evolution of actual capacity provided at Canarias ACC .......................................................37

Figure 4‐10: Monthly ATFM en‐route delay 2014 and capacity planning (Canarias) ............................37

Figure 4‐11: Monthly ATFM en‐route delay and hourly traffic profile (Athens) ....................................38

Figure 4‐12: Capacity planning at Reims ACC ........................................................................................38

Figure 4‐13: Monthly ATFM en‐route delay and hourly traffic profile (Reims) .....................................39

Figure 4‐14: Monthly ATFM en‐route delay in 2014 (Brest & Marseille) ..............................................39

Figure 4‐15: Changes in constraining ACCs (2008‐2014) .......................................................................40

Figure 4‐16: Factors affecting flight efficiency .......................................................................................42

Figure 4‐17: Free route airspace development (Map 2014) ..................................................................43

Figure 4‐18: Free route airspace implementation in 2014 by ACC ........................................................43

Figure 4‐19: Europe‐wide horizontal en‐route flight efficiency (2011‐2014) ........................................44

Figure 4‐20: Differences flight efficiency by State and FABs (2014) ......................................................45

Figure 4‐21: Share of total actual additional distance by FAB (2014) ....................................................45

Figure 4‐22: Actual additional distance per flight by FAB (2014) ..........................................................46

Figure 4‐23: Impact of the Ukrainian crisis on horizontal flight efficiency ............................................46

Figure 4‐24: Ratio of time airspace was used vs. allocated (pre‐tactically) ...........................................47

Figure 4‐25: Criteria to effective cooperation/ coordination between civil and military stakeholders 49

Figure 4‐26: ATFM performance (network indicators) ..........................................................................50

Figure 5‐1: Traffic variation at the top 30 European airports (2014/2013) ...........................................53

Figure 5‐2: Average daily operational cancellations in 2014 (24 airports included) .............................54

Figure 5‐3: Operational cancellations at the top 30 airports in 2014 ....................................................54

Figure 5‐4: ANS‐related efficiency at airports ........................................................................................56

Figure 5‐5: Average inbound and outbound delays at the top 30 airports in 2014 ..............................56

Figure 5‐6: Peak declared arrival capacity and service rate ...................................................................58

Figure 5‐7: ANS‐related inefficiencies on the arrival flow at the top 30 airports in 2014 .....................59

Figure 5‐8: Airport contribution towards European network performance (arrival flow) ....................60

Figure 5‐9: Performance drivers at London Heathrow airport ..............................................................60

Figure 5‐10: Capacity‐related ATFM regulations at Zurich airport ........................................................61

xi

Figure 5‐11: Traffic and arrival ATFM delays at Istanbul Sabiha Gökçen airport (2008‐2014) ..............61

Figure 5‐12: Average additional taxi‐in time (2014) ..............................................................................62

Figure 5‐13: ATFM slot adherence at airports (2014) ............................................................................64

Figure 5‐14: ANS‐related inefficiencies on the departure flow at the top 30 airports in 2014 .............65

Figure 5‐15: Airport contribution towards European network performance (Departure flow) ............65

Figure 5‐16: Additional Taxi‐out times at Rome Fiumicino airport ........................................................66

Figure 5‐17: Impact of winter operations on additional taxi times .......................................................66

Figure 6‐1: SES States (RP1) and non‐SES States in RP1 ........................................................................69

Figure 6‐2: Real en‐route unit costs per SU for EUROCONTROL Area (€2009) ........................................70

Figure 6‐3: Difference between 2013 and 2012 costs by nature (€2009) ................................................70

Figure 6‐4: 2013 Real en‐route ANS costs per SU by charging zone (€2009) ...........................................72

Figure 6‐5: En‐route costs and service unit growth (RP1 SES States and non‐SES States) ....................73

Figure 6‐6: Real en‐route ANS costs per SU, 2013 Actuals vs. Forecasts (in €2009) ................................74

Figure 6‐7: 2013 Real en‐route ANS costs per SU: Actuals vs. Forecasts (in €2009) by charging zone ....74

Figure 6‐8: Pan‐European en‐route cost‐efficiency outlook 2015‐2019 (in €2009) .................................76

Figure 6‐9: Geographical scope of terminal ANS cost‐efficiency analysis .............................................77

Figure 6‐10: Real terminal ANS unit costs (€2009) for reporting States ..................................................78

Figure 6‐11: Comparison of 2013 terminal ANS unit costs by TCZ (SES States (RP1)) ...........................79

Figure 6‐12: Comparison of 2013 terminal ANS actual costs vs. 2013 forecasts (Nov. 2012) ...............80

Figure 6‐13: 2013 Terminal ANS actual costs vs. 2013 forecast costs at State Level.............................80

Figure 6‐14: Real terminal ANS costs per TNSU, total costs (€2009) and TNSUs .....................................81

Figure 6‐15: Change in real terminal ANS total costs 2015‐2019 (real €2009) ........................................81

Figure 6‐16: Breakdown of gate‐to‐gate ATM/CNS provision costs 2013 (€2013) ................................82

Figure 6‐17: Conceptual framework for the analysis of economic cost‐effectiveness ..........................83

Figure 6‐18: Changes in economic cost‐effectiveness, 2009‐2013 (€2013) ..........................................84

Figure 6‐19: Changes in economic cost‐effectiveness by ANSP, 2012‐2013 (€2013) ............................84

Figure 6‐20: Changes in ATM/CNS provision costs and traffic, 2012‐2013 (€2013) ..............................85

Figure 6‐21: ATM/CNS cost‐effectiveness comparisons, 2009‐2013 (€2013) .......................................86

Figure 6‐22: Breakdown of changes in cost‐effectiveness, 2012‐2013 (€2013) ....................................86

Chapter 1: Introduction

PRR 2014‐ Chapter 1: Introduction 1

PART I- BACKGROUND

1 Introduction

1.1 Purpose of the report

The purpose of this Performance Review Report (PRR 2014) is to provide policy makers and ANS stakeholders with objective information and independent advice concerning the performance of European Air Navigation Services (ANS) in 2014, based on research, consultation and information provided by relevant parties. This is because ANS are essential for the safety, efficiency and sustainability of civil and military aviation, and to meet wider economic, social and environmental policy objectives.

PRR 2014 has been produced by the independent Performance Review Commission (PRC) which, with its supporting unit the Performance Review Unit (PRU), provides independent advice on ANS performance, and proposes recommendations, to the EUROCONTROL States. The PRC’s purpose is “to ensure the effective management of the European Air traffic management system through a strong, transparent and independent performance review”, per Article 1 of its Terms of Reference [Ref. 1]. More information about the PRC is given on the inside cover page of this report.

1.2 Key events in 2014

The tragic loss of MH17 with no survivors, in Ukrainian airspace in July, was undoubtedly the most significant event in European airspace during 2014.

Other notable events in 2014 included:

January: industrial action across Europe on 29‐31 January. This caused 70 000 minutes of delay and hundreds of flight cancellations; Georgia became the 40th EUROCONTROL Member State on 1 January 2014;

March: Industrial action in France generated 100 000 minutes of delay. Additional industrial action in Brest ACC on 20 March also generated delays;

April: Reopening of the upper airspace in Kosovo for civilian overflights on 3 April;

May: French industrial action generated significant delays in the five French ACCs;

June: Industrial action in France and Belgium caused severe network disruption: half a million minutes of delay, 2 500 fewer flights and an extra 348 000 miles flown. On 05 June and 10 June Austria, Germany, the Czech Republic and Slovakia experienced loss of secondary radar contact with several aircraft;

August: the Bárðarbunga volcano showed signs of imminent eruption, which did not occur;

September: Industrial action by Air France and Lufthansa pilots led to significant numbers of flight cancellations. Industrial action by Italian air navigation services on Saturday 6 September impacted local operations in Karlsruhe, Maastricht and Reims ACCs.

October: strong winds due to remnants of Hurricane Gonzalo during the middle of the month caused high delays particularly at London/Heathrow and Amsterdam/Schiphol.

October: Industrial action by Lufthansa pilots on 20‐21 October, and by Germanwings pilots on 15 October, resulted in the cancellation of approximately 1 650 flights.

November: The new Bosnia Herzegovina ACC opened successfully with no ATFM delay due to airspace changes in the region.

December: Industrial action at Lufthansa in early December resulted in around 1 500 fewer flights. Industrial action in Belgium on 8 and 15 December and Italy on the 12 December led to relatively few delays but it is estimated that hundreds of flights were cancelled.

PRR 2014 ‐ Chapter 1: Introduction

2

1.3 Consultation with stakeholders on PRR 2014

Communication and consultation are major priorities for the PRC. The draft final report was made available to stakeholders for consultation and written comment from 27 February‐20 March 2015. The PRC considered every comment received, and amended the final report where appropriate.

The PRC’s recommendations can be found in the Executive Summary.

1.4 Structure of the report

PRR 2014 consists of two parts:

Part I provides a consolidated high‐level view of four ANS Key Performance Areas (KPA) in the wider context of European General Air Traffic. The four KPAs are: Safety, Capacity, Environment and Cost‐efficiency. Part I also provides an overall economic evaluation of ANS performance.

Chapter 1: Introduction

Chapter 2: ANS in European Air Transport

Part II gives a more detailed analysis of ANS performance by KPA.

Chapter 3: Safety

Chapter 4: Operational En‐route ANS Performance (Capacity/Environment)

Chapter 5: Operational ANS Performance at Airports (Capacity/Environment)

Chapter 6: ANS Cost‐efficiency

1.5 Report scope

Unless otherwise indicated, PRR 2014 refers to ANS performance in the airspace controlled by the 40 Member States of EUROCONTROL1 (see Figure 1‐1), hereinafter referred to as “Europe”.

Figure 1‐1: EUROCONTROL States (2014)

The data cited in PRR 2014 relates to the calendar year 2014, with the following exceptions:

1 Estonia became the 41st Member State on 01 January 2015. It is thus outside the temporal scope of PRR 2014.

EUROCONTROL 2014

PRR 2014 ‐ Chapter 1: Introduction

3

Safety data (chapter 3): some of the 2014 data are still provisional;

Cost‐efficiency data (chapter 6): these data relate to the calendar year 2013 which is the latest year for which actual financial data are available.

1.6 Implementation status of PRC recommendations

The PRC tracks the follow‐up of the implementation of its recommendations, in accordance with Article 10.7 of its Terms of Reference [Ref. 1].

The EUROCONTROL States, at the 41st Session of the Provisional Council, (May 2014) accepted, unamended, all of the PRC’s recommendations contained in PRR 2013 [Ref. 2].

The recommendations and the associated implementation status are detailed in Annex I on page 88 of this report.

1.7 PRC as Performance Review Body of the Single European Sky

Since 2010, EUROCONTROL through its PRC supported by the PRU has been designated [Ref. 3] by the European Commission as the Performance Review Body (PRB) of the Single European Sky (SES). The PRB Chairman is appointed separately and is not a member of the PRC.

The PRB advises and assists the European Commission in the implementation of the SES performance scheme through:

Advising on the setting of performance targets;

assessing the associated performance plans at State/Functional Airspace Block level, and;

monitoring system performance under various key performance areas and indicators.

The PRC uses common procedures, tools and data for the SES performance scheme and for the EUROCONTROL performance review system. These synergies avoid overlaps between the two systems and help to ensure consistency. The PRC’s goal is to contribute to the improvement of the performance of pan‐European air navigation services, in the interests of all stakeholders.

During 2014, the SES States undertook preparatory work for the 2nd Reference Period of the SES performance scheme (RP2: 2015‐2019).

Chapter 2: ANS in European Air Transport

PRR 2014 ‐ Chapter 2: ANS in European Air Transport 4

2 ANS in European Air Transport

KEY POINTS KEY DATA 2014

After the decrease between 2011 and 2013, average daily IFR flights in Europe increased by +1.7% in 2014 with notable regional variations.

Following the continuing increase in aircraft size and average flight distance, en‐route service units (+5.9%) grew notably stronger than flights in 2014.

With the exception of Albania, Armenia, Malta, Moldova and Ukraine, all States showed a positive growth in 2014.

The airspace closure following the downing of Malaysian Airlines (MH17) in Ukraine in July 2014 had a notable impact on traffic flows shifting flights from Ukraine to adjacent States.

Compared to 2013 (which was the best year on record), arrival punctuality in Europe decreased slightly from 84.0% to 83.7% in 2014.

Based on the latest available cost projections for 2014, estimated total economic ANS‐related costs in the SES area are estimated to increase by 2.4% compared to 2013. The increase is mainly driven by the projected increase in en‐route and terminal ANS costs in 2014 and the notable increase in en‐route ATFM delays in 2014.

Traffic demand & Punctuality 2014 change vs.

2013

IFR flights controlled2 9.60M +1.7%

Flight hours controlled2 14.6M +2.5%

Total distance in km3 9 307M +3.8%

En‐route Service Units2 130.1M +5.9%

Arrival punctuality (% of flights arriving within 15 min..

after their schedule) 83.7% ‐0.3%pt.

Economic evaluation (M€ 2009) ‐ (SES area)

Projected total ANS costs (en‐route + terminal)

7 555 +3.4%

Estimated cost of inefficiencies in the gate‐to‐gate phase4

2 010 ‐2.9%

Estimated cost of en‐route and airport ATFM delay

710 +7.6%

Total estimated ANS‐related economic costs (M € 2009)

10 275 +2.4%

2.1 Introduction

This chapter provides a high‐level view of ANS performance in the wider context of air traffic operating under Instrument flight rules (IFR) in Europe. It addresses the key performance areas Safety, Cost‐efficiency (ANS costs), Capacity (ATFM delays), and Environment (flight efficiency).

After an overview of the evolution of European IFR traffic, the chapter provides a synthesis of key findings from the more detailed analyses of ANS performance in Chapters 3‐6 in order to provide an overall economic evaluation of ANS performance in Europe at the end of the chapter.

2.2 European Air Traffic Demand

After the decrease in the previous two years, average daily IFR flights increased by +1.7% in 2014, which is between the baseline (+1.2%) and the high (+2.3%) traffic scenario predicted for the ESRA 085 area in the STATFOR 7‐year forecast, published in February 2014 [Ref. 4].

For 2015, the STATFOR 7‐year forecast (2015 [Ref. 5]) expects European flights to grow by 1.5% in the baseline scenario. The average annual growth rate between 2014 and 2021 is forecast to be at +2.5%.

Figure 2‐1 shows the actual evolution of European IFR flights since 1990 together with selected traffic forecasts. As a result of the economic crisis in 2008, traffic forecasts have been continuously revised downwards and traffic is only expected to reach pre‐economic crisis levels (2008) by 2017.

2 EUROCONTROL Statistical Reference Area (ESRA) 2008 (see Glossary). 3 ESRA08 States, excluding Ukraine (see Glossary). 4 The level of inefficiency is measured with respect to reference values which, in view of necessary (safety) or

desired (capacity) limitations, are not achievable at system level. Hence, the ANS‐related inefficiencies cannot be reduced to zero.

5 The EUROCONTROL Statistical Reference Area (ESRA) is designed to include as much as possible of the ECAC area for which data are available from a range of sources within EUROCONTROL (see also Glossary).

PRR 2014 ‐ Chapter 2: ANS in European Air Transport

5

Figure 2‐1: Evolution of European IFR flights (1990‐2021)

Compared to the baseline scenario in the February 2008 forecast, there were almost three million IFR flights less in Europe in 2014 than predicted before the economic crisis.

The evolution of key traffic indices6 in Figure 2‐2 also illustrates the effects of the crisis starting in 2008. Whereas the number of flights remains below 2008 levels in 2014, there has been an increase in average aircraft size (avg. weight) and passenger numbers. This suggests that airlines responded to the crisis with a reduction in the number of services but with, on average, larger aircraft.

Figure 2‐2: European traffic indices (2004‐14)

The continuous increase in passenger numbers shown in Figure 2‐2 is also confirmed by the Association of European Airlines (AEA). According to initial estimates, AEA member airlines achieved a record load factor of nearly 81% in 2014 (79.9% in 2013) [Ref. 6].

It is also noteworthy to point out that en‐route service units (+5.9%) grew notably stronger than IFR flights (+1.7%) in 2014 as a result of the increase in average aircraft weight and longer flights.

En route service units

A service unit is used for charging purposes. It is the multiplication of the Aircraft Weight Factor (based on maximum take‐off weight) by the Distance Factor (distance in chargeable airspace).

What is also interesting to look at is the evolution of European IFR traffic by market segment between 2004 and 2014, as shown in Figure 2‐3.

6 Please note that the individual indices can refer to slightly different geographical reference areas.

‐6%

‐4%

‐2%

0%

2%

4%

6%

8%

10%

5

6

7

8

9

10

11

12

13

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

2020

% annual growth (bars)


STATFOR (Feb. 2015)7‐year forecast

( before 1997, estimation based on Euro 88 traffic variation)

IFR Flights in 2014: 9.60 M (+1.7%)

Source : EUROCONTROL/STATFOR (ESRA2008)

Feb. 2008

Feb. 2011forecast

100

105

110

115

120

125

130

135

140

145

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Inde

x 10

0 =

200

4

Sources: ACI; STATFOR (ESRA2008); CRCO

+ 2.5% Flight hours

+ 1.7% IFR flights

+ 1.7% Avg. weight (MTOW)

+ 6.1% En route Service Units

+ 5.4% Passengers (ACI)

+ 3.8% Distance

change vs. 2013 (%)


6

Figure 2‐3: IFR flights by market segment (2004‐2014)

Indeed, although traditional scheduled traffic still accounts for the majority of traffic (54%), a continuous decrease in market share since 2004 is notable. At the same time, the share of low cost carriers increased from 10.6% in 2004 to 26.9% in 2014. The other market segments such as business aviation and cargo remained relatively stable below the 10% mark over time.

European traffic growth

Figure 2‐4 shows the number of average daily flights in 2014 by State (bottom of chart) and the changes compared to 2013 in absolute and relative terms. The figure is sorted according to the absolute change compared to the previous year. Information at ACC level can be found in Annex II on page 91 of this report.

Figure 2‐4: Traffic variation by State (2014/2013)

In absolute terms, Bulgaria, Turkey, Romania and Hungary experienced the highest year‐on‐year growth in 2014 (left side of Figure 2‐4 ). With the exception of Turkey, where a significant share of the growth originates from international and domestic traffic, the growth in the aforementioned States is mainly driven by an increase in overflights following changes in traffic flow patterns in 2014.

65.7%54.0%

10.6%

26.9%

0%

10%

20%

30%

40%

50%

60%

70%

8.0

8.5

9.0

9.5

10.0

10.5

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

market share (%

)


Evolution of market segments over time (STATFOR classification)

TraditionalScheduled

Low‐Cost

Charter

BusinessAviation

Cargo

Other (incl.military)

Source: EUROCONTROL: STATFOR


7

Those changes with a notable impact on traffic growths in the States concerned were caused by a number of foreseen and unforeseen events in 2014:

April 2014: Following the re‐opening of the Kosovo sector (new Bosnia Herzegovina ACC), FYROM, Bosnia‐Herzegovina, and Serbia & Montenegro gained traffic.

April/July 2014: The Ukrainian crisis had a major impact on traffic flows in 2014. The closure of the Crimean airspace (Simferopol' FIR) in April 2014 and the closure of the airspace at the Eastern border of Ukraine (Dnipropetrovs'k FIR) following the downing of Boeing 777 MH17 in July 2014 generated many re‐routings in that region.

The flows mostly impacted by the re‐routing avoiding totally or partially Ukraine are North‐West Europe to Middle East or Asia/Pacific and Russia to Egypt (shifting traffic from Ukraine, Moldova, Azerbaijan and Armenia to Bulgaria, Turkey and other adjacent states).

July 2014: The closure of Libyan airspace reduced air traffic over Malta as aircraft avoided the central Mediterranean corridor.

The main effects of the abovementioned events are shown in Figure 2‐5 which compares actual IFR traffic growth in 2014 to the forecast growth rates in the STATFOR 7‐year forecast (February 2014) to illustrate some interesting points to note.

Four of the five largest States (Germany, Italy, UK, and Spain) experienced a traffic growth close to the high traffic forecast scenario in 2014 (left side of Figure 2‐5);

Due to a strong traffic growth in summer 2014, traffic growth in Greece, Cyprus and Turkey was higher than predicted in the high forecast scenario;

There was a notable shift of traffic as a result of the airspace closure in Ukraine. While traffic in Ukraine and Moldova dropped by ‐36.9% (on average 500 daily flights less than in 2013) and 24.2% respectively, traffic in Bulgaria (24.1%), Romania (16.6%), Hungary (11.6%), Turkey (11.2%), and Slovakia (9.8%) increased far beyond forecast levels; and,

Malta showed also a notable drop in IFR flights in 2014 (‐6.8%) as a result of the airspace closure in Libya.

Figure 2‐5: Actual IFR traffic growth vs forecast growth by State

‐36.9%

‐24.2%

24.1%

16.6%

11.6%

9.8%

‐6.8%

‐10%

‐5%

0%

5%

10%

15%

20%

25%

Greece

Cyprus

Turkey

Ukraine

Moldova

Bulgaria

Romania

Hungary

Slovakia

Malta

IFR traffic growth (%)

Actual growth(2014)

STATFORforecast(Feb. 2014)baseline(including highand low range)

Strong summer Ukraine crisis Libya

Source: STATFOR; PRC analysis

‐2%

‐1%

0%

1%

2%

3%

4%

France

Germany

Italy

UK

Spain

ESRA08

IFR traffic growth (%)

Close to high forecast range in most of the 5 largest States


8

European traffic outlook (2015‐2021)

The traffic outlook over the next seven years (see Figure 2‐6) continues to show a contrasted picture between the mature markets in Western Europe struggling to recover from the economic crisis and the emerging markets in Central & Eastern Europe for which a substantial growth is foreseen between 2014 and 2021.

Figure 2‐6: Forecast traffic growth 2014‐2021

European traffic characteristics

Traffic variability is a factor that needs to be taken into account in ATM performance review. If traffic is highly variable, resources may be underutilised, or made available when there is little demand. Variability in traffic demand is therefore likely to have an impact on productivity, cost‐efficiency, service quality and predictability of operations.

Variability can be broadly characterised as seasonal variability (difference in traffic level between different times of the year), temporal variability (difference in traffic levels between different times of the day), and spatial variability (variability of demand within a given airspace).

Different types of variability require different types of management practices, processes, and training to ensure that an ANSP can operate flexibly in the face of variable traffic demand. To a large extent, variability can be statistically predictable, and therefore adequate measures to mitigate the impact of variability could in principle be planned (for example, overtime, flexibility in breaks, and flexibility to extend/reduce shift length).

Figure 2‐7 compares the peak day for each year to the average daily number of flights.

Figure 2‐7: Peak day vs. avg. day traffic (Europe)

16%

18%

20%

22%

24%

26%

28%

30%

32%

20

22

24

26

28

30

32

34

36

2006

2007

2008

2009

2010

2011

2012

2013

2014

Average daily flights ('000) Difference

Peak vsAvg. (%)

Averagedaily IFRflights

Peak dayIFR flights



9

In 2014, the peak day was 25.2% higher than the average day but still below the peak day in 2008.

It is interesting to note that there was a notable increase in the difference between peak and the average day between 2011 and 2014.

Seasonal variability is particularly difficult for an ANSP to adapt to, as working practices that are practically feasible have only a limited ability to deal with high seasonal variability.

Figure 2‐8 provides an indication of seasonality by comparing the average weekly traffic to the peak week in 2014. The European core area shows only a moderate level of seasonality (at high traffic levels) whereas a high level of seasonal variability linked to holiday traffic is observed in South East Europe.

Figure 2‐8: Seasonal traffic variations at ATC‐Unit level (2014)

Figure 2‐8 also clearly shows the impact of the Ukrainian crisis in terms of traffic variability as a result of traffic shifting to adjacent States following the closure of the airspace at the Eastern border of Ukraine (Dnipropetrovs'k FIR) as a result of the downing of Boeing 777 MH17 in July 2014.

Traffic complexity is generally regarded as a factor to be considered when analysing ANS performance.

In 2005, a composite measure of “traffic complexity” combining traffic density (concentration of traffic in space and time) and the intensity of potential interactions between traffic (structural complexity) was developed together with interested stakeholders (see grey box).

Structural complexity and adjusted density are independent. Traffic in an area could be dense, but structurally simple; equally, traffic could be structurally complex but sparse.

Traffic complexity

The complexity score in this report is a composite measure which combines a measure of traffic density (concentration of traffic in space and time) with structural complexity (structure of traffic flows) [Ref. 7].

The structural complexity is based on the number of potential horizontal, vertical or speed interactions between aircraft in a given volume of airspace (20x20 nautical miles and 3.000 feet in height).

For example, a complexity score of 8 corresponds to an average of 8 minutes of potential interactions with other aircraft per flight hour in the respective airspace.

CAN

LIS

ANK

SCO

SHA

TAMSTO

MAD

BRE

BOD

ROM

DNI

WARLON

ISTBAR

BUC

MAA

MAR

ATH+MAK

KIE

MAL

STA

MAL

NIC

BOR

SEV

BRI

ODE

COP

SOF

LVO

ZAG BEO

OSL

TBI

RIG

VIL

BUD

MIL

TAL

WIE

RHE PRAMUNREI

PAD

RHE

TIR

PAR

CHIBRAT

ZURYERGEN

SKO

LJU

PAR

PAR

LAN

BREM

AMS

BRU

DUB

LON TC

PAL

Lower Airspace

Lower Airspace

Traffic variability 2014< 1.15

> 1.15> 1.25> 1.35> 1.45


10

The relationship between “traffic complexity” and ATM performance in general, is not straightforward. High density can lead to a better utilisation of resources but a high structural complexity entails higher ATCO workload and potentially less traffic.

Figure 2‐9 shows the evolution of complexity in Europe between 2006 and 2014. At European level, the composite measure is largely driven by the adjusted density and shows an increasing trend between 2009 and 2014.

In 2014, complexity at European system level increased to 6.56 minutes of interactions per flight hour compared to 6.26 interactions per flight hour in 2013. Figure 2‐9: Complexity trend over time (Europe)

However, at local level7 the picture is more contrasted and the complexity scores differ significantly. Figure 2‐10 shows structural complexity and the adjusted density at ANSP level in 2014. ANSPs with the highest overall complexity score are located in the top right corner. The two enable to better understand the underlying drivers of the overall complexity score.

Figure 2‐10: Structural complexity and adjusted density at ANSP level (2014)

In 2014, Skyguide traffic complexity score is ranked as the highest in Europe, followed by NATS, DFS, and Belgocontrol. A description of the methodology used to compute the complexity score and a table with the complexity scores can be found in Annex III on page 92 of this report.

7 The complexity score represents an annual average. In areas with a high level of seasonal variability the

complexity score may be higher during peak months.

5.99

6.23

6.44

6.03

6.08

6.15

6.17

6.26

6.56

0

2

4

6

8

2006

2007

2008

2009

2010

2011

2012

2013

2014

Complexity scores Overall

complexityscore

Adjusteddensity

Structuralcomplexity


2 4 6 8 10 12

0.4

0.6

0.8

1.0

1.2

1.4

Complexity scores 2014

Adjusted density

Str

uct

ura

l In

de

x

Skyguide (CH)

NATS (Continental) (UK)

DFS (DE)

Belgocontrol (BE)

MUAC (MUAC)

LVNL (NL)

ANS CR (CZ)

Austro Control (AT)

Slovenia Control (SI)

DSNA (FR)

DHMI (TR)

LPS (SK)

ENAV (IT)

HungaroControl (HU)Croatia Control (HR)

SMATSA (LY)

ENAIRE (ES)

BULATSA (BU)

PANSA (PL)

ROMATSA (RO)

SAKAERONAVIGATSIA (GE)

DCAC Cyprus (CY)

NAVIAIR (DK)

Albcontrol (AL)

M-NAV (MK)

LFV (SE)

HCAA (GR)

EANS (EE)

NAV Portugal (Continental) (PT)

LGS (LV)Oro Navigacija (LT)

Avinor (Continental) (NO)

IAA (IE)

UkSATSE (UA)

Finavia (FI)

MATS (MT)

MoldATSA (MD)

MATS (AM)

EUROCONTROL Avg.


11

2.3 Safety

Worldwide, airline safety improved to the best level on record with a rate of one fatal accident per 2.38 million flights in 2014 [Ref. 8].

Safety is clearly the primary objective of ANS. However, not all accidents can be prevented by ANS (technical failure, etc.) and there are a number of accidents without ANS involvement.

Figure 2‐11 shows the total commercial air transport (CAT) accidents8 between 2003 and 2014 (grey bars) including those accidents with ANS contribution9 (blue bars) for the EUROCONTROL area.

After the increase between 2009 and 2011, total CAT accidents decreased again between 2011 and 2013 to reach the lowest level on record in 2013. In 2014, total CAT accidents increased again slightly.

Accidents with ANS contribution (blue bars) are generally rare compared to total CAT accidents (grey bars).

Over the past five years, there were only two accidents where ANS was a factor in the causal chain of events.

Figure 2‐12 shows the total serious incidents between 2003 and 2014 (grey bars), including those with ANS contribution (blue bars) in the EUROCONTROL area.

The profile is similar with a steady decrease in serious incidents between 2010 and 2013, followed by a slight increase in 2014.

While overall the picture is positive, in view of the rare occurrence of accidents with ANS contribution, a meaningful review of ANS safety performance requires a more in‐depth analysis of ANS‐related incidents and of the effectiveness of the ANS system in place to prevent accidents and incidents in the future.

This is provided in Chapter 3 of this report.

8 Commercial Air Transport is defined by ICAO as “aircraft operations involving the transport of passengers, cargo

or mail for remuneration or hire. 9 Accidents with ANS contribution means that at least one ANS factor was in the causal chain of events leading to

the occurrence encountered by the aircraft.

Figure 2‐11: Accidents in EUROCONTROL area with ANS contribution (2003‐14)

4 3 3 1 2 1 2 1 105

1015202530354045

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Number of accidents

Source: EASA Accidents with ANS contribution

Total commercial air transport (CAT) accidents and accidents with ANS contribution (fixed wing, weight > 2250Kg MTOW)

Figure 2‐12: Serious Incidents in EUROCONTROL area with ANS contribution (2003‐14)

13 19 23 20 217 6

2011 11 8 5

0

20

40

60

80

100

120

140

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Number of serious incidents

Source: EASA Serious incidents with ANS contribution

Serious incidents in commercial air transport (CAT) (fixed wing, weight > 2250Kg MTOW)


12

2.4 Service quality

This section presents a synthesis of operational air transport performance in order to provide an estimate of the ANS‐related10 contribution towards service quality in Europe. The underlying analytical framework is described in more detail in Annex IV on page 94 of this report.

Air Transport Punctuality (airspace user perspective)

Figure 2‐13 shows arrival and departure punctuality in Europe between 2004 and 2014. Compared to 2013 (which was the best year on record), arrival punctuality in Europe decreased slightly from 84.0% to 83.7% in 2014.

Figure 2‐13: European On time performance (2004‐14)

Punctuality/ On time performance

“Punctuality” is the share of flights arriving/ departing within 15 minutes after the scheduled arrival/ departure time (airline schedules).

There are many factors contributing to the on time performance of a flight.

Punctuality is therefore the “end product” of complex interactions between airlines, airport operators, the European Network Manager and ANSPs, from the planning and scheduling phases up to the day of operation. Network effects also have a strong impact on air transport performance.

Arrival punctuality (red line) is clearly driven by departure punctuality (blue line) with only comparatively small changes once the aircraft has left the departure stand.

Although “punctuality” is a valid high level indicator from a passenger point of view for service quality, it is important to note that, from an operational viewpoint, flights arriving/departing ahead of schedule may have a similar negative effect on the utilisation of resources (i.e. TMA capacity, gate availability etc.) as delayed flights.

The variability of operations determines the level of predictability and has an impact on airline scheduling (see grey box) but also on the provision of ATC and airport capacity (i.e. TMA capacity, en‐route capacity, gate availability, etc.).

The lower the predictability the more difficult it is to match capacity to demand without inefficiencies in terms of delay (insufficient capacity) or cost (underutilisation of resources).

In order to achieve a satisfactory punctuality level, airlines frequently include time buffers in their schedules (see grey box). Generally speaking, the higher the variability, the wider the distribution of actual travel times the more difficult it is for airlines to build reliable schedules resulting in lower utilisation of resources (e.g. aircraft, crews) and higher overall costs.

10 In this report, “ANS‐related” or “ANS‐actionable” means that ANS has a significant influence on the operations.

83.7%

74%

76%

78%

80%

82%

84%

86%

88%

90%

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

% of flights

Departures within 15 min. after the scheduled departure time (%)

Arrivals within 15 min. after the scheduled arrival time (%)

Source: CODA

Departure and arrival punctuality in Europe

Airline scheduling

Airlines build their schedules for the next season on airport slot allocation, crew activity limits, airport connecting times, and by applying a quality of service target to the distribution of previously observed block‐to‐block times (usually by applying a percentile target to the distribution of previously flown block times).

The level of “schedule padding” is subject to airline strategy and depends on the targeted level of on‐time performance.


13

Figure 2‐14: Variability of flight phases (2008‐14)

‐10

‐5

0

5

10

15

20

25

2008

2009

2010

2011

2012

2013

2014

2008

2009

2010

2011

2012

2013

2014

2008

2009

2010

2011

2012

2013

2014

2008

2009

2010

2011

2012

2013

2014

2008

2009

2010

2011

2012

2013

2014

Departure time Taxi‐out phase Flight phase (en‐route &terminal)

Taxi‐in phase Arrrival time

minutes

Source: CODA; PRC Analysis

Gate-to-gate phase

Standard deviation

Range (80th-20th percentile)

Variability on intra‐European flights (2008‐2014)Figure 2‐14 gives an indication of the level of variability11 from the airspace users’ point by phase of flight. Only intra‐European flights were included to remove the impact of the jet stream on intercontinental flights which generally show a higher level of variability.

The analysis confirms that arrival times are mainly driven by variations encountered at the departure airport, with comparatively small variations in the gate‐to‐gate phase (taxi‐out, en‐route, and taxi‐in). Although, at system level, the variability in the gate‐to‐gate phase is small, it should be noted that this may vary significantly by airport or route (see chapter 5).

It should be noted that improvements in actual flight time distributions do not automatically result in improved punctuality levels, as the airline schedules for the new season are likely to be reduced by applying the punctuality target to the set of improved flight times (see box on airline scheduling).

In order to also evaluate possible changes in flight durations published in airline schedules, Figure 2‐15 shows the evolution of scheduled block times together with arrival delays.

Using the DLTA12 methodology illustrates trends over time by showing deviations relative to the average over the full analysis period.

By definition the arrival delay (blue line) will be the inverse of the arrival punctuality in Figure 2‐13.

Figure 2‐15: Evolution of delays and block times (2004‐14)

Scheduled block times on the other hand remained relatively stable over time which suggests that there was no substantial growth of schedule buffers over the period at system level.

11 In order to limit the impact from outliers, variability is measured as the difference between the 80th and the 20th

percentile for each flight phase. Flights scheduled less than 20 times per month are excluded. 12 The Difference from Long‐Term Average (DLTA) metric is designed to measure changes in time‐based (e.g. flight

time) performance normalised by selected criteria (origin, destination, aircraft type, etc.) for which sufficient data are available. It provides a notion of the relative change in performance over time but does not enable to identify the underlying performance driver.

‐4‐3‐2‐1012345

Jan‐2004

Jan‐2005

Jan‐2006

Jan‐2007

Jan‐2008

Jan‐2009

Jan‐2010

Jan‐2011

Jan‐2012

Jan‐2013

Jan‐2014

Jan‐2015Scheduled block time

Arrival delay

Deviation from the long term

average in m

inutes

Source: CODA; PRC Analysis

Intra European flights (12 months trailing average)


14

Figure 2‐13 and Figure 2‐14 show clearly that departure delays at origin airports are the main contributor towards arrival punctuality.

In order to better understand the underlying drivers, Figure 2‐16: Departure delays by cause (2008‐2014) provides a breakdown of causal factors for departure delay in Europe, reported by airlines (see grey box).

The grouping of the delay codes enables a focus on ANS‐related delays where ANS is either the root cause for the delay (i.e. ATC capacity, staffing, ATC equipment) or where an imbalance between capacity and demand (i.e. weather, etc.) was handled by ANS.

The departure delays can be further

categorised into primary (delay cause is directly attributable) and “reactionary” delays (carried over from previous flight legs).

Unsurprisingly, reactionary

delay (44.5%), caused by delay which could not be absorbed on subsequent flight legs, is the largest single delay group.

Delays due to turn round issues caused by airlines, airport operators, ground handlers and other parties accounted for 37% of all departure delays in 2014.

The ANS‐related share (ATFM en‐route, ANS related airports, and ATFM weather) in total departure delays was 13.3% in 2014. If only primary delays (all delays but reactionary delay) are concerned, the ANS‐related share in 2014 was 24.0% which represents a slight decrease to 2013 (24.1%)13.

Figure 2‐17 shows the sensitivity of the air transport network to primary delays.

In 2014, the ratio decreased slightly to 0.80 which means that, on average, every minute of primary delay resulted in some 0.80 min. of reactionary delay.

Altogether, around three quarters

13 Note that 45% of the reported ANS related delay was due to ATC pre departure delays (IATA code 89).

Departure delays

Departure delays in this report are measured compared to airline schedule. They are experienced at the stand before the aircraft departs and reported by airlines to CODA according to a set of delay codes defined by IATA. For a better focus on the ANS‐related delays the IATA delay codes were grouped:

En‐route ATFM (IATA codes 81,82);

ANS‐related airport delays (Code 83,89);

ATFM due to weather (Code 73, 84);

Weather non ATFM such as snow removal or de‐icing (Codes 71,72,76,76,77);

Reactionary delays (Codes 91‐96); and,

Local turn‐round delays: Primary delays caused by non‐ANS related stakeholders (all other Codes).

Figure 2‐17: Sensitivity to primary delays

0.80

0.4

0.5

0.6

0.7

0.8

0.9

1.0

2006

2007

2008

2009

2010

2011

2012

2013

2014ratio reactionary to primary

delay

Sensitivity of the European air transport network to primary delays

Primary delay includes local turnaround delays and en‐route and airport ATFM delays

Source: CODA analysis; PRUSource: CODA analysis; PRUSource: CODA analysis; PRU

Figure 2‐16: Departure delays by cause (2008‐2014)

44.5%

37.0%

6.6%

4.5%

0

2

4

6

8

10

12

14

16

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

2008

2009

2010

2011

2012

2013

2014

average dep

arture delay per flight

departure delay by cause (%)

Avg. delay per flight

Reactionary

Turn round relateddelays (local):

Weather (nonATFCM)

ATFM (Weather)

ANS related(airports)

ATFM (En‐route)

Source. CODA analysis; PRU


15

of the primary delays in Europe are not related to ANS and the improvement of overall air transport performance in Europe requires a thorough evaluation of all delay causes which is beyond the scope of this report14. More research would be needed to better understand the ANS contribution towards reactionary delays (which is the largest single delay group) and to identify possible ANS strategies to mitigate the propagation effects in the network.

ANS‐related service quality (service provider perspective)

This section provides a synthesis of ANS‐related efficiency and an estimate of its total impact on airspace users’ operations in terms of time and fuel burn (see also framework in Annex IV). The respective performance indicators are discussed in more detail in the corresponding chapters on operational en‐route ANS performance (Chapter 4) and ANS performance at airports (Chapter 5).

Inefficiencies in the various flight phases (airborne vs. ground) have a different impact on airspace users in terms of predictability, fuel burn (engines‐on vs. engines‐off) and costs (see also Annex IV).

For ATFM delays at the gate, fuel burn is quasi nil but with a low level of predictability before the day of operations which consequently impacts on punctuality.

ANS‐related inefficiencies in the gate‐to‐gate phase are generally more predictable as they are often related to congestion or structural issues (route design) which show similar patterns every day.

Although those “inefficiencies” in the gate‐to‐gate phase are usually considered in the airline scheduling and therefore have only a limited impact on punctuality, the impact in terms of additional fuel burn (see grey box on airline scheduling on page 12), CO2 emissions15, and associated costs is significant.

Previous research [Ref.9] suggests that the share of CO2 emissions which can be influenced by ANS is approximately 6% of the total aviation related CO2

emissions in Europe or around 0.2% of total European anthropogenic CO2 emissions (see grey box).

The next section provides an overview of the current best estimate of the total additional time, fuel burn and CO2

emissions16 that can be influenced by ANS (ANS‐related17).

Share of ANS‐related CO2 emissions

In Europe, aviation accounts for approximately 3.5% of total CO2 emissions [Ref. 10].

Analysis in previous PRRs showed that approximately 6% of the aviation‐related CO2 emissions can be influenced by ANS.

In terms of total European CO2 emissions the share that can be influenced by ANS is therefore approximately 0.2% (6% x 3.5% ≈ 0.2%).

The underlying data is derived from and should be read in conjunction with the analyses provided in Chapters 4 to 6 of this report. For the interpretation of the results in the following sections, it is worth recalling that:

the computations refer to a theoretical optimum which, due to necessary (safety) or desired (capacity) limitations, is not achievable at system level. Hence, the ANS‐related inefficiencies cannot be reduced to zero as a certain level of delay is necessary and sometimes even desirable if a system is to be run efficiently without underutilisation of available resources;

a clear‐cut attribution between ANS and non‐ANS related factors is often difficult in a

14 The Central Office for Delay Analysis (CODA) publishes detailed monthly, quarterly, and annual reports on more

delay categories (see http://www.eurocontrol.int/coda). 15 The emissions of CO2 are directly proportional to fuel consumption (3.15 kg CO2 /kg fuel). 16 It does not consider emissions from facility management (heating etc.) or ANS staff travel to/from airports which

is also relevant from an environmental point of view. 17 ANS‐related means that ANS has a direct influence in managing the performance. It does not mean that observed

inefficiencies can be reduced to zero.

≈6%

≈3.5%

Share of aviation related CO2 emissions (Europe)

Share of aviation emissions actionable by ANS

Total anthropogenic

CO2 emissions in Europe

Estimated share of ANS‐related CO2 emissions in Europe (2011)


16

complex interrelated environment such as air transport. Whereas ANS can significantly help to improve performance in the measured areas, there are inevitably factors and trade‐offs from other areas and/or stakeholders which impact on overall performance; and,

the high level overview of total ANS‐related inefficiencies combines effects from traffic variations and changes in performance. For more detailed analysis in each area, please consult the respective chapters in the report.

Figure 2‐18 provides an overview of the current best estimate of operational inefficiencies, where ANS can provide a contribution to improve performance. It is useful to recall that the total ANS‐related impact is affected by genuine performance changes but also by changes in traffic volumes.

As the relevant data are presently not available for all EUROCONTROL States at the same quality, the temporal and geographical scope of the high‐level overview is limited to SES States18.

Figure 2‐18: Estimated ANS‐related impact on fuel burn/environment

Compared to 2013, en‐route ATFM delay showed a notable increase (+14.4%) in 2014. At the same time, ANS‐related performance in the gate‐to‐gate phase (additional taxi‐out time, en‐route flight efficiency, and additional ASMA time) showed a decrease of around 4% compared to 2013.

Despite the traffic growth and the increase in en‐route ATFM delays in 2014, total additional operating time and fuel burn decreased in 2014.

Inefficiencies in the vertical en‐route flight profile, TMA departure phase, and the taxi in phase are presently not included in Figure 2‐18. The magnitude can change by region or airport and it is acknowledged that there is scope for future improvement in those areas as well as a need to include them in future estimations of ANS‐related inefficiencies in order to get an even more complete picture.

However, just as there are facets of the inefficiency pool not covered, there are system constraints and interdependencies that would prevent from achieving the theoretical optimum.

18 Due to changes in data (more complete and higher quality data) and scope (SES States only), the figures in this

section are not directly comparable to the analyses in previous PRRs. All time series analyses included in this report are based on the same data sources and methodologies and therefore comparable.

2013 2014 % change 2013 2014 % change 2013 2014

3.50 M 3.42 M ‐2.2% ‐ ‐ ‐ ‐

4.85 M 5.55 M 14.4% ‐ ‐ ‐ ‐

17.7 M 16.9 M ‐4.0% 0.3 Mt 0.2 Mt ‐6.1% 0.8 Mt 0.7 Mt

17.2 M 16.4 M ‐4.7% 0.7 Mt 0.7 Mt ‐0.3% 2.2 Mt 2.2 Mt

8.5 M 8.2 M ‐4.1% 0.3 Mt 0.3 Mt ‐0.5% 1.0 Mt 1.0 Mt

51.8 M 50.5 M ‐2.4% 1.3 Mt 1.3 Mt ‐1.5% 4.0 Mt 4.0 Mt


Estimated operationa l inefficiencies where ANS can haveEstimated tota l addi tional

(numbers may not add up due to rounding):

an impact (time, fuel burn and CO2 emiss ions ) Fuel burn

SES Performance Scheme

CO2 Operating time (min.)

ANS‐related

inefficiencies

At stand

Total estimated ANS‐related impact

Airport ATFM arriva l delay

En‐route ATFM delay

Gate‐to‐

gate

Addi tiona l taxi ‐out time (77 RP1 a i rports )

Horizonta l en‐route fl ight effi ciency (actua l )

Addi tiona l ASMA time (39 RP1 ai rports )


17

2.5 Economic evaluation of ANS performance

According to the latest available data from the Association of European Airlines (AEA), the share of air navigation charges in total operating cost in remained at around 6% in 2013 (see Figure 2‐19).

The most dominant cost factor driving airline operating expenses is fuel (24%), followed by Station & Ground, and Maintenance.

The relative share in Figure 2‐19 therefore depends a lot on changes in jet fuel prices and provides only an indication of the relative share of air navigation charges in total airline operating costs in 2013.

It is furthermore important to note that, depending on the airline business model, the share of navigation charges in operating costs can vary (i.e. low‐cost airlines may have a higher share).

Figure 2‐19: Share of ANS charges in airline operating costs (2013)

Figure 2‐20 depicts the long‐term trend of en‐route ANS costs per kilometre together with traffic evolution and en‐route ATFM delay per flight (summer) between 1990 and 2013. Different from the total economic evaluation in the next section ‐ which also considers terminal ANS performance but which is limited to 2011‐2014 ‐ Figure 2‐20 focuses on en‐route performance for which a consistent set of data is available over the entire period.

Figure 2‐20: Long‐term trend of traffic, unit costs and en‐route ATFM delay (summer)

The long term analysis shows cyclic patterns with periods of high delays and lower unit costs and vice versa until 2003 largely linked to a reactive management of ATC capacity. As of 2003, the situation improves following a more proactive approach to capacity planning combined with continuous ANS performance monitoring.

Over the past 20 years, a continuous reduction of en‐route ANS costs per kilometre is visible with en‐route ATFM delays reaching the lowest level on records in 2013.

The economic evaluation of ANS performance in this section is an attempt to monetarise direct ANS costs (en‐route and terminal) and the indirect costs due to ANS‐related inefficiencies (ATFM delays,

0.0

1.0

2.0

3.0

4.0

5.0

6.0

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

€2009 per kilometre

En‐route unit costs (€2009/km)

Traffic in km (index: 1990=1)

En‐route delay (summer)

All States in Route Charges system

Source : EUROCONTROL/CRCO, NM (delay)

Min

utes

of

en-r

oute

AT

FM

del

ay p

er f

light

and

traf

fic in

dex


18

additional taxi‐out and ASMA time, horizontal en‐route flight efficiency)19 which are both borne by airspace users in Europe. The estimation combines the high‐level cost‐efficiency results from chapter 6 with the results from the service quality chapters (4‐5).

Whilst it is not deemed appropriate to include a monetary value for safety in the economic assessment, its primacy is fully recognised.

The analysis does not consider costs for on‐board equipment nor does it provide a full societal impact assessment which would include, for instance, also the cost of delay to passengers and environmental costs. Due to data availability the analysis is restricted to SES States and the period 2011 to 2014.

The direct ANS en‐route and terminal costs were derived from Chapter 6 of this report, where a more detailed analysis is available. It is important to point out that the 2011‐2013 ANS cost figures represent actuals whereas the 2014 ANS cost figures were based on the latest available cost projections (P) and might change.

ANS‐related inefficiencies in operations impact on airspace users in terms of cost of time and fuel. Estimating the costs of such inefficiencies is a complex task requiring expert judgement and assumptions based on published statistics and the most accurate data available. There are inevitably margins of uncertainty which need to be taken into account for the interpretation of the results.

The costs of ANS‐related additional time in this report is based on a study from the University of Westminster [Ref. 11]. More information on the applied methodology for the economic evaluation of ANS performance is available in Annex V on page 96 of this report.

Fuel price is a major cost driver (see also Figure 2‐19), especially in the context of increasing jet fuel prices over the past years. The cost of ANS‐related additional fuel burn is based on EUROCONTROL models and the average jet fuel price provided by IATA.

The goal of the evaluation is to monitor ANS performance over time. Hence, in order to remove any bias from fuel price changes, the average jet fuel price in 2014 was consistently applied to all years20.

Figure 2‐21: Estimated total economic ANS‐related costs (SES States)

Figure 2‐21 shows the estimated total economic ANS‐related costs for the SES States between 2011

19 The costs of cancellations are not considered in the assessment of total economic ANS costs. 20 The “real” cost therefore might have been higher or lower in the individual years, depending on how the 2014

price compares to the price in the respective year.

2011 (A) 2012 (A) 2013 (A) 2014 (P)2013 vs .

2012

2014 vs

2013

9.3 M 9.1 M 8.9 M 9.1 M ‐1.3% 1.9%

€ 5 970 M € 6 050 M € 5 950 M € 6 155 M ‐1.6% 3.5%

€ 1 460 M € 1 410 M € 1 355 M € 1 400 M ‐3.9% 3.3%

€ 475 M € 325 M € 275 M € 270 M ‐14.5% ‐2.2%

€ 845 M € 455 M € 385 M € 440 M ‐15.5% 14.4%

€ 725 M € 685 M € 655 M € 625 M ‐4.1% ‐4.6%

€ 1 080 M € 1 005 M € 955 M € 935 M ‐4.7% ‐2.4%

€ 515 M € 485 M € 460 M € 450 M ‐4.5% ‐2.3%

€ 11 070 M € 10 415 M € 10 035 M € 10 275 M ‐3.5% 2.4%

Additiona l taxi ‐out time (77 RP1 a i rports)

En‐route extens ion (actual tra jectories )

Additiona l ASMA time (39 RP1 a i rports )

Estimated tota l ANS‐related economic costs

Service quality Ai rport ATFM delays (SES area)

En‐route ATFM delays (SES area)

Al l costs relate to States subject to the SES

Performance Scheme and are expressed in M € 2009

IFR fl ights (M)

ANS

costs En‐route ANS costs (SES area)

Terminal ANS costs (SES area)

Costs of ANS‐related inefficiencies

The estimated airline delay costs in the University of Westminster study [Ref. 11] include direct costs (fuel, crew, maintenance, etc.) the network effect (i.e. cost of reactionary delays) and passenger related costs.

Whilst passenger ‘value of time’ is an important consideration in wider transport economics, only those costs which impact on the airline’s business (rebooking, compensation, market share and passenger loyalty related costs) were included in the estimate. Estimates of future emissions costs from the EU emission trading scheme from 01 January 2012 were not included.


19

and 2013 and the provisional trend for 201421, based on the latest available ANS cost projections (see also Chapter 6).

In the single European sky area, actual en‐route and terminal ANS costs in 2013 were lower than initially projected in last year’s report (see also chapter 6) suggesting that ANSPs were able to adjust their costs to the ‐1.3% decline in traffic in 2013. Combined with the notable decrease in service quality related costs in 2013, total economic costs decreased by ‐3.5% compared to 2012 (initial projection for 2013 was +0.9%).

Based on the latest available cost projections for 2014, estimated total economic ANS‐related costs in the SES area are estimated to increase by 2.4% compared to 2013. The increase is mainly driven by the projected increase in en‐route and terminal ANS costs in 2014 and the notable increase in en‐route ATFM delays in 2014.

2.6 Conclusions

After the decrease between 2011 and 2013, flights in Europe increased again by +1.7% in 2014 with a positive medium term outlook. According to the latest STATFOR 7‐year forecast (2015), flights are expected to grow by 1.5% in 2015 and to continue with an annual average growth rate of 2.5% between 2014 and 2021.

However, despite the positive growth in 2014 and the promising outlook, the impact of the economic crisis on the industry is still prominent. At European level, there were almost three million flights less in 2014 than initially predicted before the economic crisis in 2008.

Whereas the number of flights still remains below 2008 levels in 2014, there has been a continuous increase in average aircraft size and passenger numbers during the period which suggests that airlines responded to the crisis with a reduction in the number of services but with, on average, larger aircraft. Hence, as a result of the increase in average aircraft weight and longer flights, en‐route service units (+5.9%) grew notably stronger than flights (+1.7%) in 2014.

With the exception of Albania, Armenia, Malta, Moldova and Ukraine, all EUROCONTROL states showed a positive growth in 2014, with Turkey remaining the main contributor to the growth in Europe. Four of the five largest States (Germany, Italy, UK, and Spain) experienced a traffic growth close to the high forecast scenario in 2014 and traffic was even higher than predicted in the high forecast scenario in Greece and Cyprus.

European traffic was affected by a number of foreseen and unforeseen events in 2014. The tragic loss of MH17, in Ukrainian airspace in July was undoubtedly the most significant event in European airspace during 2014. As a result, there was a notable shift of traffic. While traffic in Ukraine and Moldova dropped by ‐36.9% and ‐24.2% respectively, traffic in Bulgaria (24.1%), Romania (16.6%), Hungary (11.6%), Turkey (11.2%), and Slovakia (9.8%) increased far beyond forecast levels but could be accommodated without noteworthy delays.

Worldwide, airline safety improved to the best level on record with a rate of one fatal accident per 2.38 million flights in 2014. In the EUROCONTROL area safety levels remain high although preliminary 2014 data suggests a slight increase in total commercial air transport accidents and ANS‐related incidents in 2014 following the lowest level on record in 2013.

Compared to 2013 (which was the best year on record), arrival punctuality in Europe decreased slightly from 84.0% to 83.7% in 2014. Reactionary delay, caused by delay which could not be absorbed on subsequent flight legs, remains the largest single delay group (44.5%) in 2014, followed by delays due to turn round issues (37%). The ANS‐related share (en‐route and airport combined) in total departure delays was 13.3% in 2014.

21 Due to changes in data (more complete and higher quality data) and scope (SES States only), the figures in this

section are not directly comparable to the analyses in previous PRRs. The time series included in this report are based on the same data sources and methodologies and therefore comparable.


20

Although the decrease in arrival punctuality in 2014 was mainly driven by a deterioration in turn‐round performance at airports, en‐route ATFM delays also increased compared to 2013. Due to this increase in en‐route ATFM delay in 2014 and an increase in flights at slightly better efficiency levels (taxi‐out, en‐route, terminal) than in 2013, estimated total additional ANS‐related operating time increased by 2.4% in 2014.

The total economic evaluation of ANS performance presents a consolidated view of direct ANS costs and estimated indirect ANS‐related costs (ATFM delays, additional taxi‐out and ASMA time, horizontal en route flight efficiency) borne by airspace users.

In the Single European Sky (SES) area, actual en‐route and terminal ANS costs in 2013 were lower than initially projected in last year’s PRR, suggesting that ANSPs were able to adjust their costs to the ‐1.3% decline in traffic in 2013. Hence, combined with the notable lower delay related costs observed in 2013, total economic costs decreased by ‐3.5% compared to 2012 (initial projection for 2013 was +0.9%).

Based on the latest available ANS cost projections for 2014, estimated total economic ANS‐related costs in the SES area are estimated to increase by 2.4% compared to 2013. The increase is mainly driven by the projected increase in en‐route and terminal ANS costs and the notable increase in en‐route ATFM delays in 2014.

Chapter 3: Safety

PRR 2014 ‐ Chapter 3: Safety 21

PART II – KEY PERFORMANCE AREAS

3 Safety

KEY POINTS KEY DATA (2013)


Over the past four years (2011‐2014) there were no fatal ANS‐related accidents and only one accident with ANS contribution in 2013.

The number of serious incidents with ANS contribution continued the positive trend observed since 2010 and decreased further in 2014 to the lowest level on record.

Separation minima infringements remained the single largest category for serious ANS‐related incidents, followed by runway incursions.

Despite the considerable progress over the past years there is still a need to further improve safety data quality and completeness.

Performance indicators (AST reporting)

2013 % change vs. 2012

Total number of reported separation min. infringements

2 097 + 16.8%

Separation minima infringements (Severity A+B)

262 ‐ 10.0%

Total number of reported runway incursions

1 421 + 15.2%

Total number of reported runway incursions (A+B)

75 + 53.1%

Total number of reported unauthorised penetration of airspace

3 435 ‐ 31.4%

Unauthorised penetration of airspace (Severity A+B)

38 ‐ 36.7%

3.1 Introduction

This chapter reviews the Air Navigation Services (ANS) safety performance of the EUROCONTROL Member States between 2004 and 2014.

Sections 3.2 and 0 in this chapter show the trends in ANS‐related accidents and incidents in the EUROCONTROL area. Section 3.4 provides an analysis of the current status of safety data reporting and investigation in EUROCONTROL Member States while Section 3.5 outlines potential benefits of aviation risk modelling for future performance improvements.

The review of ANS‐related accidents and incidents in this chapter is based on:

Accident and serious incidents from data contained in the EASA database; and,

Incident data from 2004 to 2013 and preliminary 2014 data reported to EUROCONTROL via the Annual Summary Template (AST) reporting mechanism.

The scope of the safety review in this chapter is summarised in Figure 3‐1 below.

Analysis scope Type Category Weight

Accident

(EASA DB)

ANS related

ANS contribution Commercial Air Transport (CAT) Fixed wing >2 250 Kg

Serious Incidents

(EASA DB)

ANS related

ANS contribution CAT Fixed wing >2 250 Kg

Incidents

(EUROCONTROL AST) ANS related All All

No limitation

Figure 3‐1: Sources and scope of ANS safety review in this chapter

Note that final investigation reports for some accidents and incidents might be delayed by more than two years, particularly when the investigation is complex. For this reason, it may be possible that the historic results shown in this report differ slightly from previous Performance Review Reports. In addition, the scope of the review may be changed in future reports depending on the added value for reviewing the ANS safety performance and on the improvement in data granularity and data quality.

PRR 2014 ‐ Chapter 3: Safety

22

3.2 Accidents and serious Incidents (ANS‐related and with ANS contribution)

Accidents (ANS‐related and with ANS contribution)

Figure 3‐2 shows accidents (ANS‐related and with ANS contribution) in Commercial Air Transport (CAT) involving fixed wing aircraft with more than 2250kg in the EUROCONTROL area between 2004 and 2014.

ANS‐related accidents decreased notably between 2007 and 2013 but increased again in 2014.

ANS‐related vs. ANS contribution

“ANS‐related” means that the ANS system may not have had a contribution to a given occurrence, but it may have a role in preventing similar occurrences in the future.

“ANS contribution” means that at least one ANS factor was in the causal chain of events leading to an occurrence, or at least one ANS factor potentially increased the level of risk, or it played a role in the occurrence encountered by the aircraft.

Figure 3‐2: Accidents (ANS‐related and with ANS contribution) in the EUROCONTROL area

Over the past four years (2011‐2014) there were no fatal ANS‐related accidents (dark blue bars) and only one accident with ANS contribution in 2013 (brown line).

Figure 3‐3 shows the cumulative number of ANS‐related accidents (2012‐2014) by occurrence category22. It should be noted that some accidents may have been assigned to more than one occurrence category.

Figure 3‐3: ANS‐related accidents by occurrence category (EUROCONTROL

area)

The majority of ANS‐related accidents between 2012 and 2014 were related to adverse weather (TURB) and ground collisions (GCOL).

22 As defined by ICAO Commercial Aviation Safety Team taxonomy (CAST/ICAO).

012345678910

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Number of accidents

ANS‐related accidents and accidents with ANS contribution in Commercial Air Transport (CAT), (fixed‐wing, weight > 2 250 kg)

Fatal

Non‐Fatal

Accidents withANS Contribution

Source: EASA

7 41 1 1 1 1

0

2

4

6

8

Turbulence

(TURB)

Ground Collision

(GCOL)

Cabin Safety

Even

ts (CABIN)

Evacuation

(EVAC)

Runway

excursion (RE)

Abnorm

alRunway Contact

(ARC)

Windshear or

thunderstorm

(WSTRW)

Number of acciden

ts

Cumulative (2012 + 2013 + 2014) number of ANS‐related accidents by occurrence category

(CAT, fixed wing, weight > 2 250 kg)

Source: EASA


23

Serious incidents (ANS‐related and with ANS contribution)

Figure 3‐4 shows ANS‐related serious incidents23 and serious incidents with ANS contribution in CAT (fixed‐wing aircraft > 2 250kg) in the EUROCONTROL area between 2004 and 2014.

Figure 3‐4: Serious incidents (ANS‐related and with ANS contribution) in the EUROCONTROL area

In the EUROCONTROL area, the number of ANS‐related serious incidents decreased continuously between 2010 and 2013 but showed an increase in 2014. At the same time, the number of serious incidents with ANS contribution (brown line) continued to decrease, so that 2014 had the lowest number of such incidents in the eleven year period.

Figure 3‐5 shows the cumulative number of ANS‐related serious incidents by occurrence category (taxonomy per CAST/ICAO) between 2012 and 2014. Note that only the top 5 categories are shown and that some of the serious incidents might have been assigned to more than one category.

As in previous years, near Mid‐Air Collision, i.e. loss of separation in the air (Near MAC), Runway Incursions (RI‐VAP), and ATM/CNS continue to be the most frequent ANS‐related serious incidents between 2012 and 2014.

Near MAC = Near Mid‐Air Collision

RI‐VAP = Runway Incursions Vehicle, Aircraft, Person

ATM/CNS = Air Traffic Management / Communication Navigation Surveillance

WSTRW = Windshear or thunderstorm

CFIT = Controlled Flight Into Terrain

Figure 3‐5: ANS‐related serious incidents by occurrence category (EUROCONTROL area)

23 A serious incident is defined as an incident involving circumstances indicating that an accident nearly occurred.

0

10

20

30

40

50

602004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014N

umber of serious incidents

ANS‐relatedserious incidents

Seriousincidents withANScontribution

Source: EASA

Serious incidents (ANS‐related and with ANS contribution) in Commercial Air Transport (fixed‐wing, weight> 2 250kg)

32 15 135 4

0

10

20

30

Near M

AC

RI‐VAP

ATM

/CNS

WSTRW

CFIT

Number of serious inciden

ts

Cumulative (2012 + 2013 + 2014) number of ANS‐related serious incidents by occurrence category

(top 5) (CAT, fixed‐wing, weight > 2 250 kg)

Source: EASA


24

3.3 ATM‐related incidents

This section provides a review of ATM‐related incidents, reported through the EUROCONTROL Annual Summary Template (AST) reporting mechanism. The PRC has made use of, with gratitude, the data provided by the EUROCONTROL Safety Regulation Commission (SRC) in its Annual Reports 2013 and 2014 [Ref. 12] and in its Intermediate Report 2015. The data for 2014 is only preliminary with updates foreseen for the September 2015 reporting session. The commentary in this section therefore relates to the final 2013 figures.

The “Severity A” categorisation in the EUROCONTROL AST corresponds to “Serious incident” in the EASA database. The number of “Severity A” incidents in the AST is still slightly higher than the total “Serious incidents” in the EASA database. The reasons for this difference may be related to the criteria used by the Safety Investigation Authorities (SIAs) for selecting serious incidents and by the notification procedures and practices24 used at national level for notifying about Severity class A.

The inconsistency issue, together with other issues related to the quality and completeness of safety occurrence data (AST, ECR and EASA database), is constantly closely monitored by a Task Force composed of members of the Performance Review Unit (PRU), EASA and the EUROCONTROL Directorate Pan‐European Single Sky (DPS).

Airspace ‐ Separation Minima Infringements

In 2013, the total number of reported Separation Minima Infringements (SMIs) increased by almost 17%, compared to the previous year. This increase is in line with the trend started in 2011.

Figure 3‐6 depicts the number of reported risk‐bearing SMIs (Severity A and B) in EUROCONTROL airspace between 2004 and 2014(P).

Figure 3‐6: Reported high‐risk SMIs in EUROCONTROL States (2004‐14P)

After a continuous increase between 2009 and 2012, risk bearing SMIs decreased again in 2013 to 12% of the total number of reported SMIs. Overall, there was a decrease in the total number of serious incidents for both risk bearing Severity categories in 2013.

24 These issues have also been identified in a number of States during ICAO USOAP audits.

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 20132014(P)

N° of States reporting 29 26 26 28 28 29 30 31 33 36 36 40

Total n° reported 889 1 226 1 281 1 398 1 567 1 711 1 418 1 402 1 571 1 796 2 097 2 274

Severity B 152 164 243 250 295 236 141 178 217 258 232 243

Severity A 76 67 80 73 70 56 27 16 35 33 30 21

% : Proportion of Severity A+B 26% 19% 25% 23% 23% 17% 12% 14% 16% 16% 12% 12%

76 67 80 73 70 5627 16

35 33 30 21

152 164

243 250295

236

141 178

217258

232 243

0%

5%

10%

15%

20%

25%

30%

0

100

200

300

400

Number of occurren

ces

Separation Minima Infringements

Severity B Severity A % : Proportion of Severity A+B

Source: SRC Intermediate Report 2015


25

Airspace ‐ Unauthorised Penetration of Airspace

The total number of Unauthorised Penetrations of Airspace (UPAs) in 2013, also known as Airspace Infringements (AIs), reported in EUROCONTROL Member States decreased by 31%, compared to 2012.

As illustrated in Figure 3‐7, the share of risk bearing (Severity A and B) UPAs, within total reported UPAs, decreased further in 2013 to 1.1%. Both risk bearing categories of UPAs show a decrease in 2013 in terms of total numbers.

Figure 3‐7: Reported high‐risk UPAs in EUROCONTROL States (2004‐2014P)

Airports ‐ Runway Incursions

Total reported Runway Incursions (RI) reported in EUROCONTROL Member States increased by approximately 15% in 2013.

Figure 3‐8: Reported high‐risk RIs in EUROCONTROL States (2004‐2014P)

In 2013, the risk‐bearing RIs (Severity A and B) increased by 53% compared to 2012 which corresponds to 5% of the total number of reported RIs (Figure 3‐8).

Both risk‐bearing categories of RIs increased in 2013. In absolute terms, serious runway incursions (Severity A) increased only marginally in 2013 (+2) whereas a more significant increase (+24) was observed in the number of major events (Severity B) reaching a level comparable to 2011.

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 20132014(P)


Total n° reported 1 178 1 209 1 983 2 041 2 416 2 797 3 336 3 381 4 742 5 010 3 435 4 288

Severity B 31 19 58 48 49 52 53 79 68 50 37 53

Severity A 28 7 16 8 5 3 6 4 12 10 1 9

% : Proportion of Severity A+B 5% 2% 4% 3% 2% 2% 2% 2% 2% 1% 1% 1%

28

716

8 5 3 6 412 10 1 9

31

19

58

48 49 5253

79 68

50

37

53

0%

1%

2%

3%

4%

5%

6%

0

20

40

60

80

100

Number of Occurren

ces

Unauthorised Penetration of Airspace



2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 20132014(P)


Total n° reported 387 564 629 680 885 926 1 094 1 385 1 399 1 234 1 421 1 439

Severity B 38 32 52 51 44 40 36 77 62 37 61 76

Severity A 25 16 9 13 12 14 15 22 23 12 14 25

% : Proportion of Severity A+B 16.3% 8.5% 9.7% 9.4% 6.3% 5.8% 4.7% 7.1% 6.1% 4.0% 5.3% 7.0%

2516

9 13 12 14 15 22 2312 14

25

38

32 5251

44 40 36

7762

37

61

76

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

0

20

40

60

80

100

120

Number of Occurren

ces

Runway Incursions




26

ATM Specific Occurrences

This section provides a review of the evolution of the risk bearing ATM specific occurrences reported through the AST, as updated in the September 2014 reporting cycle.

Figure 3‐9: Reported high‐risk ATM Spec. Occurrences in EUROCONTROL States (2004‐2014P)

ATM specific occurrences encompass those situations where the ability to provide safe ATM services is affected. Therefore, this type of occurrence typically includes failures that would affect the ANS providers’ capability to deliver safe ATM services.

The total number of reported ATM specific occurrences in 2013 showed a 7% increase compared to 2012. Risk‐bearing ATM specific occurrences showed a continuous decreasing trend since 2010. Although, the total number of all risk‐bearing ATM specific occurrences in 2013 decreased by 21% (Figure 3‐9), occurrences with Severity AA and A increased.

3.4 Reporting and Investigation

This section provides a review of the quality and completeness of ATM safety occurrences (operational and ATM specific occurrences) reported through the AST reporting mechanism, as updated in September 2014 and April 2015 (where applicable).

Total number of reported occurrences

For each EUROCONTROL Member State, the level of reporting is measured by normalising the total number of reported ATM‐related occurrences against the number of flight hours in the State. The main influencing factors for the level of reporting are the level of Just Culture and the effectiveness of the Mandatory Occurrence Reporting Systems (MORS). However, the influencing factors are not presented in this report.

The annual level ATM‐related incident reporting in Figure 3‐10 is compared to the average ECAC reporting level in 2003, which represents the brown baseline.

The number of Member States reporting above the baseline in 2013 (24) was more than three times higher than the number of States reporting under the baseline (7).

Final 2013 data was received from 36 States and the preliminary 2014 data was submitted by a record number of 40 States. In addition, arrangements have been made with Georgia, which joined EUROCONTROL in 2014 to start reporting the AST data at the earliest opportunity.

0

100

200

300

400

500

600

700

800

900

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014(P)

Number of Occurren

ces

ATM Specific Occurrences

Severity B

Severity A

Severity AA



27

Figure 3‐10: Total number of reports and level of reporting (2004‐2014P)

After a continuous increase in the total number of incidents reported by Member States between 2004 and 2012, the data submitted for 2013 showed a 17% decrease in reported ATM related incidents. A decrease in the number of reported occurrences is observed in almost all the risk categories in 2013.

The available data however does not allow drawing conclusions if the observed decrease in reported incidents in 2013 represents a genuine safety performance improvement or if it is due to diminished reporting levels. In addition, the decrease in 2013 could be partially related to the fact that several Member States indicated that they did not submit preliminary data/reports on time.

Unclassified or undetermined occurrences

Figure 3‐11 shows the number of ATM‐related incidents not severity classified or with severity classification not determined (Severity D) for different occurrences categories. The analysis is based on the data submitted via AST in September 2014, covering the reporting year 2013.

Although there has been an overall improvement compared to 2012, 24% of all reported occurrences were still not severity classified in 2013. If the occurrences where the severity is ‘not determined’ are added (i.e. some data provided but insufficient to fully assess the severity), the percentage rises to over 30%. Nevertheless this still represents an improvement when compared to the previous year’s data (43%).

SMI = Separation minima infringements

RWY = Runway incursions

IS = Inadequate separation

UAP = Unauthorised penetrations of airspace

CLR = Deviation from ATC clearance

Figure 3‐11: Severity not classified or not determined (2005‐2014P)

Considering each type of occurrence separately, the percentage varies between 7% and 51%. If “not determined” (i.e. some data provided but not enough to fully assess the severity) are also included, the range increases to 10% and 52% of total number of reported occurrences in each occurrence category.

0

2

4

6

8

10

12

14

16

18

0

5

10

15

20

25

30

35

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014(P)

Flight hours in the reporting States (million)

Number of inciden

ts reported

(thousands)

Evolution of the number of reported occurrences


Base lineECAC reporting level in 2003

0

5

10

15

20

25

30

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Sta

tes

per

cat

ego

ry

Level of reporting ATM related incidents(baseline 2003)

Less than half of the baseline

More than half of the baseline

Above baseline

Source: SRC Annual Report 2014

0

500

1 000

1 500

2 000

2 500

3 000

0%

10%

20%

30%

40%

50%

60%

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014(P)

Total number of occurences not severity

classified

% by category

Occurrences NOT Severity Classified

SMI

RWY

IS

UAP

CLR



28

An increase of the percentage of occurrences not severity classified in 2013 is reported for three types of occurrence: SMIs, UPAs, and Inadequate Separation (IS). As also indicated in the SRC 2014 Report, since the application of the Risk Analysis Tool (RAT) methodology for deriving the severity of SMIs is a key safety performance indicator of the single European sky performance scheme, action is needed to ensure this gap is closed.

In terms of numbers, the percentage of SMI occurrences not severity classified increased from 5% in 2012 to 7.4% in 2013. If the SMIs where the severity was not determined (severity class D) are added the percentage rises to 9.5%.

Conversely, some progress has been made with the severity assessment of the RIs which improved by 1% compared to 2012. However, if the situation where the severity is absent is added it amounted to almost 16%.

In conclusion, the number of unclassified or not determined incidents is still higher (over 30%) than the levels in 2006/7. As already pointed out in last year’s report, the situation is of concern, “not only for the outcome of the analysis at European level, but also for the outcome of national safety analysis (as this outcome is indicative for the results of analysis and investigation performed at national level), the sustainability of the human reporting system25, and other potential downstream repercussions such as inadequate prevention of similar incidents or inadequate sharing and dissemination of lessons learned”.

Completeness of safety data reported via the AST mechanism

Figure 3‐12 shows the typical fields that are either left blank or marked Unknown in the AST, submitted by the EUROCONTROL Member States in 2013.

ATM contribution = direct; indirect; none

Type operation = GAT or OAT

Airspace = Class of airspace: A,B,C,D,E

Flight Rules = IFR or VFR

Traffic Type = General Air Traffic, Commercial, Military

Phase of Flight = taxi, take‐off, climb to cruise, cruising, approach

Figure 3‐12: Completeness of AST reported data in 2014

The amount of fields left blank is much higher than the field where the word Unknown was inserted. It ranges from 7% up to over 75% of the reported operational occurrences. ATM contribution to the occurrence, which is the most relevant data for determining the performance of the ATM system, is left blank in case of over 7% of the reported incidents.

It is of great concern that the data required to populate a number of fields is still missing in high percentages. This lack of completeness of AST data hampers comprehensive safety analysis at European level.

25 When ATCOs or pilots provide safety reports, if feedback is not provided it can have an adverse impact on

the motivation to report.

0%

20%

40%

60%

80%

100%

ATM

Contribution

Type

operation

Airspace

Flight Rules

Traffic Type

Phase of

Flight

Completeness of the AST reported data (2013)

Empty +Unknown (%)

Empty (% )

Unknown (%)

Source: SRC Annual Report 2014


29

Figure 3‐13: States with automatic reporting tools (2014)

CHFR

BE

IT

RO

GB

BG

PT

HU

LT

GE

SK

HR

AL

OTHER TOOL + EUROCONTROL ASMT

EUROCONTROL ASMT

EUROCONTROL (HQ+EEC)

OTHER TOOL

Automatic safety data monitoring ‐ Implementation Status (2014)

In last year’s PRR, the PRC reiterated that the current manual reporting should be complemented by independent monitoring based on automatic safety data acquisition tools. Moreover, the PRC concluded that there is a need to accelerate the deployment of automatic safety data reporting tools in Europe in order to improve the reporting culture and consequently the level of reporting and recommended and encouraged States to do so.

This section provides an update of the implementation status of various automatic safety data solutions in EUROCONTROL States.

Figure 3‐13 displays the current status of deployment of the EUROCONTROL Automatic Safety Monitoring Tool (ASMT) as well as equivalent/comparable tools26 in EUROCONTROL States.

Based on information provided by the Network Manager, at the moment the ASMT is deployed and used by 14 Member States, which represents increase from the last year (+2 States vs. 2013).

As indicated in last year’s PRR, one of the widely deployed systems in Europe, for detection of potentially unsafe runway events (runway safety nets for controllers) are provided through Advanced Surface Movement Guidance & Control System (A‐SMGCS).

Similarly to last year’s observation, implementation levels of the A‐SMGCS of both Level 1 (Improved Surveillance) and Level 2 (Surveillance + Safety Nets) for 2014 indicate that additional work is still needed in order to achieve the objectives set in the ATM Master Plan by the end of 2017 (although the majority of the participating airports declared that they will achieve the objectives by the end of 2015) [Ref. 13]. Although, A‐SMGCS Level 2 may bring safety benefits, it has to be noted, that it will not enable complete automated reporting of events.

Out of 72 EUROCONTROL Member States airports, which are addressed through the ESSIP/LSSIP process (airports participating in airport implementation objectives and/or are main airports of a Member State), 46 had or have the intention to implement A‐SMGCS Levels 1 and 45 airports Level 2.

Figure 3‐14: A‐SMGCS implementation status

26 Note that information on comparable tools might be incomplete, as additional organisations may be using similar

tools that PRC was not aware of at the time of publishing of this report.

48%52%

A‐SMGCS Level 1 implementation status

Late

Completed

Source: LSSIP 2014

36%

18%

40%

A‐SMGCS Level 2 implementation status

Late

Planned

PartlyCompleted

Completed

Source: LSSIP 2014


30

Figure 3‐14 shows the implementation status of A‐SMGCS Levels 1 and 2. The analysis excludes airports which declared that this objective is “not applicable” to them. The applicability area includes only airports for which the objective is likely to deliver significant benefits.

The implementation date for Level 1 was December 2011. LSSIP 2014 states that only 52% of the 46 airports have achieved full operational capability. The 22 airports lagging behind with the implementation of A‐SMGCS Level 1 are shown in Figure 3‐15 below.

Airport Country Airport Country Airport Country

Brussels (EBBR) Belgium Milan (LIMC) Italy Barcelona (LEBL) Spain

Sofia (LBSF) Bulgaria Milan (LIML) Italy Mallorca (LEPA) Spain

Toulouse (LFBO) France Venice (LIPZ) Italy Kiev (UKBB) Ukraine

Marseille (LFML) France Rome (LIRF) Italy Manchester (EGCC) UK

Nice (LFMN) France Chisinau (LUKK) Moldova London (EGLL) UK

Düsseldorf (EDDL) Germany Warsaw (EPWA) Poland Edinburgh (EGPH) UK

Athens (LGAV) Greece Lisbon (LPPT) Portugal

Thessaloniki (LGTS) Greece Bucharest (LROP) Romania

Source: LSSIP 2014

Figure 3‐15: Airports late for A‐SMGCS Level 1 implementation

Insofar as Level 2 is concerned, the deadline for full operational capability is December 2017 (ATM Master Plan). Some 40% of the 45 airports had fully implemented Level 2 in 2014.

The use of A‐SMGCS or other equivalent runway safety programmes to improve reporting and safety performance is strongly encouraged by the PRC.

3.5 Aviation Safety Risk

For several years, the PRC is highlighting the need for a new way of representing safety performance in order to further improve ANS safety, without endangering achieved progress so far, including the level of reporting. The current methodology and system do not give a possibility to openly represent the real problems in the ANS system, as the States are protected by the fact that “benchmarking” in safety is not allowed by different legal mechanisms.

In order to improve ANS contribution to the total aviation safety in the future, the new framework should allow addressing and identifying whether or not there was a real change in performance in some of the key risk areas in Europe. This requires that the underlying data are fully made available by the States in the expected quality.

Besides a political push, to finally enable benchmarking with improved safety data, the introduction of a new approach, the development of a European concept of Acceptable Level of Safety (ALoS), and maybe even additional indicators (based for example on independent automatic data flows) will be required to show what exactly is happening to the system and what and where the real risks are.

Moreover, the Aviation System is becoming increasingly automated and more complex. Complexity is increased by not only changing equipment on the ground and in the air, but also by changing the roles of the aviation professionals. Complexity arises at the technological, work situation, organizational and political/economical levels. Traffic is once more increasing as Europe recovers from the economic downswing and this is leading to challenging safety objectives to ensure that the capacity increase is carefully managed.

For all these reasons the PRC is actively monitoring current developments on these issues and analysing its potential use for its own purposes. One of the recent ones being the work between the FAA and EUROCONTROL on the evolution of an aviation risk model (please see Annex VI for more details).


31

3.6 Conclusions


Over the past four years (2011‐2014) there were no fatal ANS‐related accidents and only one accident with ANS contribution (MET related) in 2013. The number of serious incidents with ANS contribution continued the positive trend observed since 2010 and decreased further in 2014 to the lowest level on record. Separation minima infringements remained the single largest category for serious ANS‐related incidents, followed by runway incursions. Although considerable progress has been made over the past years, there is still a need to further improve safety data quality and completeness.

After the continuous increase in reporting levels between 2004 and 2012, the total number of incidents reported fell again by 17% in 2013. However, with the available data it is difficult to differentiate whether the decrease in 2013 is due to genuine safety performance improvements or a drop in the level of reporting. Overall, final 2013 data was received from 36 EUROCONTROL Member States via the Annual Summary Template (AST) reporting mechanism and the preliminary 2014 data was submitted by a record number of 40 States.

Although already flagged up as a crucial area for improvement in previous PRRs, the share of unclassified incidents, as reported through the AST mechanism, remained unacceptably high in 2013 even though improvements are visible. There is an urgent need for further action in order to closely monitor trends and to support the prevention of similar incidents. Lastly, more effort should be put in place by the Member States to ensure that the number of ATM related occurrences not severity classified, continues to decrease.

The need to accelerate the deployment of automatic safety data monitoring to complement manual reporting was also already addressed in last year’s report. The latest available information from the Network Manager suggests that 14 Member States now use ASMT which represents an increase of two States compared to 2013. However, there is a concern that the implementation of A‐SMGCS Level 1 (improved surveillance) lags behind. By 2014, only 52% of the 46 airports have achieved full operational capability. Being a prerequisite for the implementation of A‐SMGCS Level 2 (Surveillance + Safety Nets), this indicates that additional work is still needed in order to achieve the objectives set in the ATM Master Plan by the end of 2017 and avoid delayed implementation of Level 2. The use of A‐SMGCS or other equivalent runway safety programmes to improve reporting and safety performance is strongly encouraged by the PRC.

To ensure that safety remains an integral part in the evolution of the more and more complex aviation system, States and service providers are encouraged to proactively engage in safety risk modelling. The improved understanding of how the various elements of the ATM system contribute to overall safety levels not only helps to better identify risk areas today but also how modifications to the system could improve performance in the future.

Chapter 4: Operational En‐route ANS Performance

PRR 2014 ‐ Chapter 4: Operational En‐route ANS Performance 32

4 Operational en‐route ANS Performance


In 2014, en‐route ATFM delay increased again for the first time since 2010. Total en‐route ATFM delays in the EUROCONTROL area grew by +17.2% in 2014 which corresponds to an average delay of 0.61 minutes per flight.

The most constraining ACCs in 2014 were Nicosia, Warsaw, Lisbon, Athinai/Macedonia, Canarias, Brest, Reims, and Marseille which accounted together for 54.6% of all en‐route ATFM delays but only 17.8% of total flight hours controlled in Europe.

En‐route flight efficiency continued to improve in 2014. At European level, filed flight plans improved from 4.86% in 2013 to 4.70% in 2014. Actual trajectories improved at a higher rate than filed flight plans from 3.14% in 2013 to 2.72% in 2014.

Review of only three Member States identified significant differences, not only between the national arrangements and the FUA specification but also between the individual Member States, in how the airspace is managed to provide the optimum benefit for civil and military airspace users.

2014 change vs.

2013

IFR flights controlled 9.6M +1.7%

Capacity: En‐route ATFM delays

Total en‐route ATFM delay (min.)

5.87M +17.2%

Average en‐route ATFM delay per flight (min.)

0.61 +0.08

Flights delayed > 15 min. en‐route (%)

1.6% +0.3%pt.

Environment: Flight inefficiency

Avg. horizontal en‐route inefficiency (flight Plan)

4.70% ‐0.16%pt.

Average horizontal en‐route inefficiency (actual)

2.72% ‐0.42%pt.

4.1 Introduction

This chapter reviews operational en‐route ANS performance. Section 4.2 reviews Air Traffic Flow Management (ATFM) delays originating from en‐route restrictions. Section 4.3 provides an evaluation of en‐route flight efficiency in Europe. Section 4.4 provides an assessment of the impact of the Ukrainian crisis on observed performance in 2014. Civil military cooperation including a review of the application of flexible use of airspace is provided in Section 4.5 before European ATFM performance at network level is addressed in the final section of the chapter.

4.2 En‐route ATFM delays

After the continuous improvement between 2010 and 2013, en‐route ATFM delays, for the EUROCONTROL area, increased by +17.2% in 2014 which corresponds to 0.61 minutes of en‐route ATFM delay per flight.

Figure 4‐1: Average en‐route ATFM delay (1997‐2014)

2.2

2.9

4.5

2.9

2.5

1.4

0.9

0.8 0.9 1.0 1.2 1.4

0.9

2.0

1.1

0.63

0.53

0.61

70

80

90

100

110

120

130

140

150

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Traffic index (base: 1997)

En‐route ATFM delay/ flight (m

in.)

ATC Capacity & Staffing ATC Other (strike, equipment, etc.)WEATHER OTHER (Special event, military, etc.)IFR Traffic

Average en route ATFM delay per flight (EUROCONTROL area)


PRR 2014 ‐ Chapter 4: Operational En‐route ANS Performance

33

Traffic flows in Europe in 2014 were affected by a number of events including the downing of Malaysian Airline MH17 in Ukrainian airspace in July, which is discussed further in Section 4.4 below.

ATC capacity and staffing‐related delays increased in 2014. They remain, by far, the main driver of en‐route ATFM delays, followed by weather and “ATC Other” which comprises, inter alia, ATC industrial actions. Delay related to ATC industrial action increased again in 2014.

The number of flights affected by ATFM en‐route delays increased again from 2.7% to 3.2% in 2014. Overall, 1.6% of flights were delayed by more than 15 minutes due to ATFM regulations, compared to 1.3% in 2013.

Figure 4‐3 shows the monthly evolution of en‐route ATFM delays and IFR flights in Europe between 2011 and 2014.

Figure 4‐3: Monthly evolution of en‐route ATFM delays (2010‐2014)

The seasonal pattern peaking together with traffic in summer is clearly visible although less pronounced over the past three years.

Local ATFM en‐route performance per ACC

While capacity constraints can occur from time to time, ACCs should not generate high delays on a regular basis. Figure 4‐4 shows the delay performance in terms of the number of days with significant en‐route ATFM delays (>1 minute per flight). As in previous years, the selection threshold for the table in Figure 4‐4 was set at greater than 30 days and the most constraining ACCs are analysed in more detail in the next sections of this chapter.

The most constraining ACCs in 2014 were Nicosia, Warsaw, Lisbon, Canarias, Brest, Athinai+Macedonia, Reims and Marseille. Together, they accounted for 54.6% of all en‐route ATFM delays but only 17.8% of total flight hours controlled in Europe.

15

17

19

21

23

25

27

29

31

33

0.0

0.5

1.0

1.5

2.0

2.5

1 2 3 4 5 6 7 8 910

11

12 1 2 3 4 5 6 7 8 9

10

11

12 1 2 3 4 5 6 7 8 9

10

11

12 1 2 3 4 5 6 7 8 9

10

11

12

2011 2012 2013 2014

avg. daily IFR flights ('000)

en‐route ATFM delay per flight

Source: PRC Analysis; Network Manager

Figure 4‐2: En‐route delay per flight by classification

8.8%

5.7%

3.4%

2.7%3.2%

5.2%

3.0%

1.7%1.3% 1.6%

0%

2%

4%

6%

8%

10%

12%

14%

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

2010 2011 2012 2013 2014

% o

f d

ela

yed

IF

R f

lig

hts

en-r

ou

te A

TF

M d

elay

per

fli

gh

t (m

in.)

ATC Capacity & Staffing ATC Other (strike, equipment, etc.)

Weather Other (Special event, military, etc.)

En route ATFM delayed flights En route ATFM delayed flight (>15min.)


34

Figure 4‐4: Overview of most constraining ACCs

Figure 4‐5 shows the evolution of ATFM en‐route delays between 2010 and 2014 at the most constraining ACCs listed above. The delay classifications, as reported by the local flow management positions (FMP), are provided and, in order to give an indication of the traffic level, the number of controlled IFR flights is plotted as a blue line.

Figure 4‐5: Delay categories for most constraining ACCs

The next section evaluates the most constraining ACCs in 2014 in more detail in order to provide a better understanding of what is affecting the performance during periods of highest delay.

Delivery of planned performance

Nicosia ACC

Capacity performance at Nicosia ACC improved from an average of 2.16 minutes per flight in 2013 to 1.91 minutes per flight in 2014. The improvement in capacity performance was with an increase in traffic levels (+9.7% on 2013 traffic levels). The number of days with an average ATFM delay of more than 1 minute per flight decreased slightly from 198 in 2013 to 191 in 2014. However, notwithstanding this improvement, Nicosia ACC remains the most constraining ACC in the network.

In 2014, 68% of en‐route delays were allocated by Nicosia FMP as being due to ATC capacity and ATC staffing, as opposed to 58% in 2013.

In April, the month of the highest delay, practically all delay was allocated to ATC Capacity and Staffing. However, April was not a month with particularly high traffic levels. Traffic levels were higher every month from May to October, with much less capacity and staffing constraints.

As reported in PRR 2013, Nicosia ACC planned to open a 5th en‐route sector during 2013. However, examination of the regulations requested by Nicosia ACC show that the delays during April were

Most constraining

ACCs in 2014

Days en

‐route

ATFM

>1 min.

En‐ro

ute d

elay

/flight (m

in.)

% of fligh

ts

delayed

>15

min.

En‐ro

ute d

elay

('000)

ATC

Capacity &

Staffing

ATC

Other

Weath

er

Other (sp

ecial

event, m

ilitary)

% of to

tal en‐

route d

elay

Traffic growth

vs 2013 (%

)

5 Year A

nnual

average growth

rate (09‐14)

% of to

tal flight

hours 2

014

Nicosia 191 1.91 5.3% 581 68% 3% 1% 28% 9.9% 9.7% 2.7% 1.0%

Warsaw 104 0.84 2.5% 547 68% 0% 8% 24% 9.3% 1.2% 4.3% 2.4%

Lisbon 61 0.53 1.6% 240 96% 2% 2% 0% 4.1% 6.9% 3.5% 2.2%

Canarias 44 0.42 1.3% 118 76% 0% 23% 1% 2.0% 6.9% 1.2% 1.2%

Brest 43 0.53 1.2% 496 46% 48% 2% 4% 8.4% 4.2% 2.6% 3.4%

Athinai+Macedonia 43 0.42 1.3% 275 94% 4% 2% 0% 4.7% 8.4% 1.0% 3.1%

Reims 40 0.42 1.2% 387 78% 7% 12% 3% 6.6% 3.8% 3.0% 1.8%

Marseille AC 34 0.57 1.2% 568 33% 59% 6% 2% 9.6% ‐0.6% 0.3% 2.7%

En‐route ATFM delay Traffic demand

0

200

400

600

800

1 000

1 200

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

2010

2011

2012

2013

2014

2010

2011

2012

2013

2014

2010

2011

2012

2013

2014

2010

2011

2012

2013

2014

2010

2011

2012

2013

2014

2010

2011

2012

2013

2014

2010

2011

2012

2013

2014

2010

2011

2012

2013

2014

Nicosia Warsaw Lisbon Canarias Brest Athinai+Macedonia

Reims Marseille

IFR flights ('000)

En‐route ATFM delay per flight (m

in.)

ATC Capacity & Staffing ATC Other (strike, equipment, etc.)

Weather Other (Special event, military, etc.)

Source: PRC Analysis; Network manager


35

experienced when only 2 or 3 en‐route sectors were open. This suggests an inability to effectively deploy existing capacity rather than a fundamental lack of additional capacity (the 5th en‐route sector has still not been implemented.)

Figure 4‐6: Monthly ATFM en‐route delay 2014 and capacity planning (Nicosia)

The Network Operations Plan (NOP) 2014‐2018/19 contains the latest capacity plans for Nicosia ACC. Figure 4‐6 clearly shows that Nicosia ACC has been postponing or downgrading capacity plans for several years. Nicosia ACC plans to have less capacity in 2019 (57) than previously promised for 2012 (62) in the capacity plans published in LSSIP 2011 for 2012‐2016.

If Nicosia ACC had implemented the (subsequently) downgraded capacity plans published in LSSIP 2012, and provided a capacity of 62 aircraft per hour in April 2014, it would have satisfied the hourly traffic demand at all times, saving airspace users approximately €10 million (129k minutes of delay @ €83 per minute) in additional costs.

Warsaw ACC

Warsaw ACC experienced a significant increase in en‐route ATFM delay in 2014 with 0.84 minutes per flight compared to 0.54 minutes per flight in 2013 (+67.74%). In 2014, Warsaw ACC had 104 days where en‐route ATFM delay was greater than 1 minute, as compared to 62 days in 2013. There was a marginal increase in traffic from 2013 to 2014 (1.2%)

Warsaw ACC allocated 68% of en‐route delays as being due to ATC capacity and ATC staffing. On the other hand, it is difficult to validate the delay contribution “Other causes”, reported between January and May 2014.

Figure 4‐7: Monthly ATFM en‐route delay and hourly traffic profile (Warsaw)

Implementation of the new PEGASUS 21 ATM system contributed significantly to an increase in delay, particularly between May and August 2014. During this period, Warsaw ACC limited all sectors to 85% capacity, (119 aircraft per hour instead of 140).

The Network Operations Plan (NOP) 2014‐2018/19 contained capacity plans for Warsaw ACC, it planned to deliver an average capacity of 140 aircraft per hour in 2014. Providing such capacity would have satisfied the hourly traffic demand 98% of the time.

0500

1,0001,5002,0002,5003,0003,5004,0004,5005,000

Jan-

14

Feb

-14

Mar

-14

Apr

-14

May

-14

Jun-

14

Jul-1

4

Au

g-14

Se

p-14

Oct

-14

Nov

-14

Dec

-14

All othercauses

Industrialaction 'I'

Weather 'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day

Monthly ATFM en-route delay and traffic (Nicosia)


40

45

50

55

60

65

70

75

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

flights per hour

Source: NOP

Development of capacity plans for Nicosia ACC

2010‐2014

2011‐2015

2012‐2016

2013‐2017

2014‐2019

0500

1,0001,5002,0002,5003,0003,5004,0004,5005,000

Jan-

14

Feb

-14

Mar

-14

Apr

-14

May

-14

Jun-

14

Jul-1

4

Au

g-14

Se

p-14

Oct

-14

Nov

-14

Dec

-14

All othercauses


Weather 'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day

Monthly ATFM en-route delay and traffic (Warsaw)


0. flts

20. flts

40. flts

60. flts80. flts

100. flts

120. flts

140. flts160. flts

180. flts

200. flts

00h0

001

h00

02h0

003

h00

04h0

005

h00

06h0

007

h00

08h0

009

h00

10h0

011

h00

12h0

013

h00

14h0

015

h00

16h0

017

h00

18h0

019

h00

20h0

021

h00

22h0

023

h00

Warsaw hourly traffic profile (June-August 2014)

MAX CTFM Traffic MIN CTFM TrafficAVG CTFM Traffic NOP Capacity


36

The improvement in capacity performance following the finalised implementation of new PEGASUS ATM system gives an encouraging indication that capacity performance in Warsaw ACC will improve for 2015 and beyond.

Lisbon ACC

Lisbon ACC experienced a significant increase in en‐route ATFM delay in 2014, with an average ATFM delay of 0.53 minutes per flight compared to 0.27 minutes per flight in 2013. There were 61 days when Lisbon ACC produced en‐route ATFM delay above one minute per flight, more than double the 26 days with en‐route delay above 1 minute per flight in 2013. In 2014, traffic levels increased by 6.9% on 2013 figures.

Lisbon ACC allocated 96% of en‐route delays as being due to ATC capacity and ATC staffing.

Figure 4‐8: Monthly ATFM en‐route delay (Lisbon)

Strikingly, the vast majority of delay occurred in November 2014, when traffic levels were significantly lower than during the summer period. Looking back at capacity performance in Lisbon ACC for the years 2011 to 2013 gives a similar picture: significant delays in the November – December period despite lower traffic levels than the spring ‐ summer periods (which are usually handled without significant delay). It would appear that there are operational constraints which only apply during this period: therefore, a targeted resolution to these specific constraints could bring considerable improvement to overall capacity performance at Lisbon ACC.

Special events:

Lisbon hosted the European Champions League Cup final in May 2014, an event that brought 200 additional movements to and from LPPT over a two‐day period, without any adverse impact on capacity in the Lisbon ACC.

0

500

1,000

1,500

2,000

Jan-

11

Feb

-11

Mar

-11

Apr

-11

May

-11

Jun-

11

Jul-1

1

Aug

-11

Sep

-11

Oct

-11

Nov

-11

Dec

-11

All othercauses


Weather'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day

2011 monthly ATFM en-route delay and traffic (Lisbon)


0

500

1,000

1,500

2,000

Jan-

12

Feb

-12

Mar

-12

Apr

-12

May

-12

Jun-

12

Jul-1

2

Aug

-12

Sep

-12

Oct

-12

Nov

-12

Dec

-12

All othercauses


Weather'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day



0

500

1,000

1,500

2,000

Jan-

13

Feb

-13

Mar

-13

Apr

-13

May

-13

Jun-

13

Jul-1

3

Aug

-13

Sep

-13

Oct

-13

Nov

-13

Dec

-13

All othercauses


Weather'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day



0

500

1,000

1,500

2,000

Jan-

14

Feb

-14

Mar

-14

Apr

-14

May

-14

Jun-

14

Jul-1

4

Aug

-14

Sep

-14

Oct

-14

Nov

-14

Dec

-14

All othercauses


Weather'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day




37

Canarias ACC

Canarias ACC experienced a marginal improvement in average en‐route ATFM delay per flight for 2014 (0.42 minutes), compared to 2013 (0.44 minutes), with a traffic increase of 6.9% over the same period. However, in 2014, there were 44 days when Canarias ACC produced en‐route ATFM delay above one minute per flight, compared to 37 days in 2013.

As was the case in 2013, the en‐route weather‐related delays for 2014 are predominantly associated to weather and runway configurations at specific airports within the FIR, and not to en‐route weather phenomena such as CB’s or Clear Air Turbulence.

Errors or inconsistencies in categorising delay, such as classifying airport related delay as en‐route delay, risks masking the real constraints to capacity, thus preventing effective mitigation or resolution.

If all the en‐route weather‐related ATFM delay in Canarias ACC was actually airport‐related delay, the average en‐route ATFM delay for 2014 would be 0.32 minutes per flight instead of 0.44 minutes per flight. In addition, the number of days where ATFM delay was more than 1 minute per flight would reduce to 37.

Traffic levels in August were similar to the traffic levels each month in the periods Jan‐May and Oct‐Dec. However, the delays due to ATC capacity were twice as much as for any other month of the year.

The Network Operations Plan 2014‐2018/2019 already flagged up concern that there would be a capacity shortfall in the Canarias ACC in 2014, and that unless the existing capacity plans are improved, capacity will lag behind demand every year until 2020.

Year 2006 2007 2008 2009 2010 2011 2012 2013 2014

Capacity 71 71 71 71 59 61 65 65 65Source: Network Operations Reports 2007–2013/ 2014‐2018/19

Figure 4‐9: Evolution of actual capacity provided at Canarias ACC

It is interesting to note that according to previous Network Operations reports, there was a sharp drop‐off in capacity between 2009 and 2010, from 71 to 59 aircraft per hour. Although available capacity grew again in 2011 and 2012 it has stalled at 65 aircraft per hour: 10% less than what was already available in 2006. It would be interesting to know what caused the reduction in capacity and what steps have been taken to regain the ‘lost’ capacity.

Figure 4‐10: Monthly ATFM en‐route delay 2014 and capacity planning (Canarias)

Athinai+Macedonia ACCs

Following real improvements in capacity performance for 2012 & 2013 (with 7 & 5 days respectively, of more than 1 minute average ATFM delay per flight), compared with 2011 (94 days), Athinai & Macedonia ACCs had 43 days when en‐route ATFM delay was above one minute per flight.

The average ATFM delay per flight for 2014 increased to 0.42 minutes per flight, up from 0.15 minute

0

100

200

300

400

500

600

700

800

900

Jan

-14

Fe

b-1

4

Mar

-14

Apr

-14

May

-14

Jun

-14

Jul-1

4

Aug

-14

Sep

-14

Oct

-14

Nov

-14

Dec

-14

All othercauses


Weather 'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day

Monthly ATFM en-route delay and traffic (Canarias)


60

65

70

75

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

flights per hour

Source: NOP

Development of capacity plans for Canarias ACC

2010‐20142011‐20152012‐20162013‐2017


38

Figure 4‐12: Capacity planning at Reims ACC

per flight in 2012 and 0.06 minutes per flight in 2013. In 2014, traffic levels increased by 8.4% on 2013 figures.

Athinai & Macedonia ACCs allocated 94.1% of ATFM delay to ATC Capacity and Staffing.

Notwithstanding the significant improvements in Athinai and Macedonia ACCs in the past three years, they are still playing ‘catch‐up’ for years when additional capacity measures were not implemented. This is especially noticeable during peak summer traffic periods June‐September, when traffic levels are more than double the demand for the winter months. This is evident by the graphs in Figure 4‐11 below showing the rise in traffic over the summer period (and accompanying delay spike) and secondly, the deployment of available capacity to handle a traffic demand (AVG CTFM Traffic) in line with the declared capacity value published in the NOP (NOP capacity).

The PRC notes the concerns expressed in the Network Operations Plan 2014‐2018/2019, about likely capacity problems in Greece for the future. The PRC can only encourage the Greek authorities to ensure that the appropriate capacity plans to handle the peak summer traffic are both developed and implemented as soon as possible.

Figure 4‐11: Monthly ATFM en‐route delay and hourly traffic profile (Athens)

Reims ACC

Reims ACC had 40 days where en‐route ATFM delay was greater than 1 minute. Traffic levels in 2014 increased by 3.8% on 2013 figures. Overall, 78% of en‐route delays were allocated by Reims ACC as being due to ATC capacity and ATC staffing in 2014.

Reims ACC did not add any additional capacity between 2008 and 2012, retaining a constant capacity of 178 aircraft per hour. In 2013 additional capacity was implemented to reach 192 per hour. However, in planning capacity for the period 2014‐2017, Reims ACC decided to provide capacity based on the low traffic forecast (2.4% AAG) rather than the reference capacity profile (5.8% AAG).

When there was a slight rise in traffic the lack of spare capacity resulted in significant delays. In the graphs below, the proximity of the AVG CTFM Traffic to the NOP capacity value highlights that the declared available capacity was deployed for significant periods.

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

Jan-

14

Feb

-14

Mar

-14

Apr

-14

May

-14

Jun-

14

Jul-1

4

Au

g-1

4

Se

p-1

4

Oct

-14

Nov

-14

Dec

-14

All othercauses


Weather 'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day

Monthly ATFM en-route delay and traffic (Athinai+Macedonia)


0. flts

20. flts

40. flts

60. flts

80. flts

100. flts

120. flts

140. flts

160. flts

180. flts

00h0

001

h00

02h0

003

h00

04h0

005

h00

06h0

007

h00

08h0

009

h00

10h0

011

h00

12h0

013

h00

14h0

015

h00

16h0

017

h00

18h0

019

h00

20h0

021

h00

22h0

023

h00

Athens hourly traffic profile (August 2014)



39

It is encouraging to note that from September the capacity situation appears to be greatly improved. The PRC hopes that Reims ACC will improve its existing capacity plans to ensure that minor increases above forecasted traffic levels do not result in significant delays for airspace users.

Figure 4‐13: Monthly ATFM en‐route delay and hourly traffic profile (Reims)

Marseille & Brest ACCs

The en‐route disruptions due to French industrial action in January, March, May and June were the principle cause of delay for both Marseille and Brest ACCs.

Figure 4‐14: Monthly ATFM en‐route delay in 2014 (Brest & Marseille)

However, the industrial action by French ATC did not just affect capacity performance in France. It had a significant impact on the capacity performance of adjacent States and on the network as a whole. For example, during the period 24/25 June, when there was an ATC strike in France, Karlsruhe UCC experienced a traffic increase of 7% compared to the previous week and delays rose by more than 19 thousand minutes; Maastricht UAC experienced a traffic rise of 4% compared to the previous week whilst delays rose by almost 32 thousand minutes; Madrid ACC despite a traffic decrease of 3% compared to the previous week allocated delays of more than 19 thousand minutes as being due to the industrial action in France.

0

500

1,000

1,500

2,000

2,500

3,000

3,500Ja

n-14

Feb

-14

Mar

-14

Apr

-14

May

-14

Jun-

14

Jul-

14

Aug

-14

Sep

-14

Oct

-14

Nov

-14

Dec

-14

All othercauses


Weather 'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day

Monthly ATFM en-route delay and traffic (Reims)


0. flts

50. flts

100. flts

150. flts

200. flts

250. flts

300. flts

00h0

001

h00

02h0

003

h00

04h0

005

h00

06h0

007

h00

08h0

009

h00

10h0

011

h00

12h0

013

h00

14h0

015

h00

16h0

017

h00

18h0

019

h00

20h0

021

h00

22h0

023

h00

Reims hourly traffic profile (May-August 2014)


0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

Jan-

14

Feb

-14

Ma

r-14

Apr

-14

May

-14

Jun-

14

Jul-1

4

Aug

-14

Sep

-14

Oct

-14

Nov

-14

Dec

-14

All othercauses


Weather 'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day

Monthly ATFM en-route delay and traffic (Marseille)


0

1,000

2,000

3,000

4,000

5,000

6,000

Jan-

14

Feb

-14

Mar

-14

Apr

-14

May

-14

Jun-

14

Jul-1

4

Aug

-14

Sep

-14

Oct

-14

Nov

-14

Dec

-14

All othercauses


Weather 'W'

Staffing 'S'

Capacity 'C'

IFR flights

aver

age

per

day

Monthly ATFM en-route delay and traffic (Brest)



40

Figure 4‐15: Changes in constraining ACCs (2008‐2014)

ACC 2008 2009 2010 2011 2012 2013 2014Nicosia 197 193 262 160 169 198 191 196Warszawa 269 225 158 75 77 62 104 139Madrid 98 96 163 168 11 10 4 79Wien 136 170 148 20 16 20 4 73Langen 27 74 143 124 71 14 18 67Karlsruhe UAC 130 78 167 47 5 11 18 65Barcelona 4 4 167 134 63 40 21 62Zagreb 138 70 89 49 19 5 27 57Canarias 33 46 79 86 25 37 44 50Athinai+Macedonia 94 54 44 94 7 5 43 49Marseille 9 5 129 53 33 25 34 41Zurich 108 83 61 15 5 5 3 40

Avg. 2008-2014

Nr. Of days with en route ATFM dly. >1min.

Changes in constraining ACCs (2008‐2014)

Figure 4‐15 shows the evolution of performance at ACCs that were considered constraining ACCs for at least three years between 2008 and 2014.

Nicosia ACC: the PRC notes that this has remained a significant problem for more than seven years without effective mitigation or resolution.

Warsawa ACC: despite significant improvement from 2008‐2009, there were still problems for airspace users in Poland in 2014. The PRC is encouraged by the improved performance towards the end of 2014 and hopes that this will bring resolution to severe capacity constraints in Polish airspace.

Madrid ACC: vast improvements in capacity performance since 2011.

Wien ACC: vast improvements in terms of capacity performance since 2010.

Langen ACC: vast improvements in capacity performance since 2012.

Karlsruhe UAC: vast improvements in capacity performance since 2011.

Barcelona ACC: improvements in capacity performance since 2011.

Zagreb ACC: improvements in capacity performance since 2011.

Canarias ACC: improvements since 2011. The PRC urges Canarias ACC to resolve the capacity problems highlighted in the chapter above, so that the capacity performance for airspace users can be further improved.

Athinai & Makedonia ACCs: The PRC notes the vast improvements in capacity performance since 2012, although it is concerned by the lack of capacity to handle the 2014 peak summer traffic. The PRC encourages the Greek authorities to ensure that capacity plans are developed and implemented so that the demands of airspace users can be satisfied in the future.

Marseille ACC: The PRC notes that the primary reason for delays at Marseille ACC since 2010 has been due to industrial action.

Zurich ACC: vast improvements in capacity performance since 2010.

In all the ACCS where there have been capacity improvements, the PRC would like thank the ACC personnel for their efforts, which are now bearing fruit for the airspace users.


41

4.3 En‐route Flight Efficiency

This section evaluates en‐route27 flight efficiency inside Pan‐European airspace. En‐route flight efficiency has a horizontal (distance) and a vertical (altitude) component.

The focus of this section is on the horizontal component, which, in general, is considered to be of higher economic and environmental importance than the vertical component across Europe as a whole [Ref. 14]. Flight efficiency at and around airports, which also includes airborne holdings, is addressed in the evaluation of ANS‐related performance at airports in Chapter 5.

The Pan‐European airspace analysed in this section refers to the Network Manager’s area of responsibility which is wider than the geographical scope of the SES performance scheme.

The analysis is based on the comparison of the actual or planned flight trajectory length with the corresponding achieved distance (see also grey box). The methodology is fully consistent with the SES Performance Scheme.

The planned trajectory is derived from the flight plans submitted by airspace users to the Network Manager. The actual flown trajectory is based on processed radar track data (Correlated Position Reports) submitted by ANSPs to the EUROCONTROL Enhanced Tactical Flow Management System (ETFMS). Work is in progress to further improve the data in terms of quality and coverage. States subject to the SES performance scheme are requested to provide radar data at a harmonised update rate of 30 seconds as of 2015, which will further enhance data quality, and all EUROCONTROL States are encouraged to also provide radar data of the same quality.

Horizontal en‐route flight efficiency

Horizontal en‐route flight efficiency compares the length of flight trajectories to the corresponding “achieved” distance.

The achieved distance apportions the Great Circle Distance (GCD) between two points within the European airspace (reference area). For the vast majority of flights, the origin and destination coincide with the airports. If the origin/ destination airport is located outside of European airspace, the entry/exit point into the reference area is used for the calculation. The indicator is calculated as the ratio of the two sums (length of trajectories and achieved distances), over all flights considered.

The methodology enables better quantification between local inefficiency (deviations between entry and exit point within a given airspace such as FAB, ANSP, ACC) and the contribution to the network (deviation from the GCD between the origin and destination airports).

The full methodology is described in more detail in the metadata which is available online [Ref. 15].

Improved flight efficiency has not only an economic impact in terms of fuel savings but also a notable environmental impact on climate in terms of reduced carbon dioxide CO2 emissions.

It is recognised that the impact of air traffic on climate is wider than just CO2 emissions. The impact also depends on emission location, time, and type of emission (CO2, water vapour, nitrogen oxides). The potential of climate‐optimised flight routing as a measure to reduce the atmospheric impact of aviation was investigated by the European REACT4C project for a sample of trans‐Atlantic flights [Ref. 16]. Such weather‐dependent flight trajectory optimisation illustrated possibilities for further reducing overall climate impact at only moderate cost increases (or even cost decreases for airlines if combined with the emission trading system). However, improved meteorological data and further research is needed before the concept can be applied at a wider scale in the future.

It is also acknowledged that the distance‐based flight efficiency indicators in this section only serve as proxies for fuel efficiency as the most fuel efficient route depends on wind. However, even the wind‐optimal route might not necessarily correspond to the choice of the airspace users because they might use different measures based on total costs (time, route charges, etc.).

The full set of information required for considering such user‐preferred alternatives (wind‐optimal, total costs, etc.) is currently not available and would require airspace users to either provide more detailed information or to agree on a standard method for the calculation of the route “values”.

27 The “En‐route” section excludes the 40 nautical mile circles around the airports (terminal areas).


42

Despite these limitations, the distance‐based flight efficiency indicators used in this section provide a consistent and stable Europe‐wide measure to identify areas for improvement and to monitor progress over time.

Factors affecting horizontal en‐route flight efficiency

En‐route flight efficiency is affected by a number of factors including:

route network design (existing route network);

route availability (utilisation of civil military structures);

flight planning capabilities (use of software, repetitive flight planning);

user preferences (time, cost, fuel);

tactical ATC routings; and,

special events such as severe weather, ATC strikes.

As indicated in Figure 4‐16, the shortest possible route is constrained by the design of the route network. Historically, the route network was structured in reference to the navigational limitations of aircraft and to enable air traffic control to provide separation with the tools available.

The shortest available route is specific to a flight and takes into account the additional constraints introduced by the Route Availability Document (RAD) and Conditional Routes (CDRs).

The RAD has the effect of modifying the route network available to specific flows of traffic, while CDRs have the effect of modifying the route network available at specific times. Both reduce the set of available routes. Further work is needed to identify the shortest route available for each flight at the time the flight commences.

The filed flight plan is the ultimate output of the planning process. Differences between the shortest available route and the route in the filed flight plan can arise because the airspace user might not be aware that the route is available, or is aware but chooses an alternative route for operational or business reasons.

As technology for both aircraft and ATC has improved, the need for such a rigid en‐route structure has diminished, to the extent that free‐route airspace (see grey box) with a positive effect on flight efficiency would now be possible throughout the entire EUROCONTROL area.

The implementation of “Free route airspace (FRA) initiatives” aims at enhancing en‐route flight efficiency with subsequent benefits for airspace users in terms of time and fuel as well as a reduction of CO2 emissions for the environment.

By the end of 2014, the Network Manager had coordinated, through the European Route Network Improvement Plan (ERNIP) [Ref. 17], the development and/or implementation of more than 200

Figure 4‐16: Factors affecting flight efficiency

Free Route Airspace (FRA) Concept

Free route airspace (FRA) is a key development with a view to the implementation of shorter routes and more efficient use of the European airspace.

FRA refers to a specified airspace within which users may freely plan a route between a defined entry point and a defined exit point, with the possibility to route via intermediate (published or unpublished) way points, without reference to the ATS route network, subject to airspace availability. Within this airspace, flights remain subject to air traffic control.

The aim of the FRA Concept Document is to provide a consistent and harmonised framework for the application of FRA across Europe in order to ensure a co‐ordinated approach.


43

airspace improvement packages relating to various FRA projects (including Night Routes and Weekend direct routings (DCTs)). These contributed to the flight efficiency improvements observed over the past years and are expected to further improve efficiency in the future.

The map below, which has been provided by the Network Manager, shows that, by the end of 2014, 30 of the 64 ACCs had implemented various steps of Free Route Operations (see Figure 4‐17), with clear benefits in terms of flight efficiency (see PRR 2013 [Ref. 2]).

Figure 4‐17: Free route airspace development (Map 2014)

Figure 4‐18 shows the ACCs that have now either fully or partially implemented Free Route Airspace operations in 2014 (Source: Network Manager):

Full Free Route Airspace implementation

Within Lisbon ACC plus NW portion of Madrid FIR (Santiago, Asturias)

Within Kobenhavn ACC, Malmo ACC and Stockholm ACC as part of SWE/DNK FAB

Within Shannon ACC/UAC as part of the ENSURE ‐ EN‐route Shannon Upper Airspace Redesign project

Full Night Free Route Airspace implementation

Within Sofia ACC

Within Chisinau ACC

Within Bucuresti ACC

Within Tampere ACC

Comprehensive DCT implementation (Night‐, Weekend‐, H24 DCTs)

Within Maastricht UAC as part of FRAM – Free Route Airspace Maastricht

Within Karlsruhe UAC as part of FRAK – Free Route Airspace Karlsruhe

Between Maastricht UAC and Karlsruhe UAC as part of FRAMaK – Free Route Airspace Maastricht and Karlsruhe (cross‐border)

Within Wien ACC

Within Zagreb, Beograd ACC AoR (including Montenegro and Bosnia & Herzegovina)

Within Skopje ACC

Within Ljubljana ACC

Within Madrid ACC (SAN and ASI sectors) as part of the FRASAI project

Within Malta ACC

Comprehensive DCT implementation (Night DCTs)

Within Milano, Padova, Roma, Brindisi and Praha ACCs

Limited DCT implementation (Night DCTs)

Within Reims, Brest, Bordeaux, Marseille ACCs and Warsaw ACCs

New Night Time Fuel Saving Routes within London, Prestwick, Part Milano, Roma and Brindisi ACCs

Figure 4‐18: Free route airspace implementation in 2014 by ACC



44

The improvement of European flight efficiency and the optimisation of the European route network is, by definition, a Pan‐European issue which requires a holistic approach carefully coordinated by the Network Manager. Uncoordinated, local initiatives may not deliver the desired objective, especially if the airspace is comparatively small and a large proportion of the observed inefficiency is due to the interface with adjacent States or FABs.

In view of the numerous factors and complexities involved, and with traffic levels growing again, flight efficiency improvements will become more and more challenging and will require the joint effort of all stakeholders coordinated by the Network Manager.

Europe‐wide en‐route flight efficiency

Figure 4‐19 shows the horizontal en‐route flight efficiency for the actual trajectory and the filed flight plan for the EUROCONTROL area. An “inefficiency” of 5% means, for instance, that the extra distance over 1 000NM was 50NM.

Figure 4‐19: Europe‐wide horizontal en‐route flight efficiency (2011‐2014)

A continuous improvement for the filed flight plan (brown line) and the actual trajectory (blue line) between 2011 and 2014 is visible (see left side of Figure 4‐19). At the European level, filed flight plan inefficiencies improved from 4.86% in 2013 to 4.70% in 2014. Inefficiencies in actual trajectories improved at a higher rate than those of filed flight plans from 3.14% to 2.72% for the same period.

Despite the further notable improvement in 2014, the stakes remain high. In 2014, the total additional distance flown compared to the reference trajectory (2.72%) was 177M kilometres. While it is important to bear in mind that the level of inefficiency cannot be reduced to zero at the system level, there is clearly scope for further improvement.

Geographical distribution of flight inefficiency

Figure 4‐20 shows the level of inefficiency in filed flight plans and in actual trajectories by Functional Airspace Block (FAB) and State in 2014.

As already pointed out in PRR 2013 [Ref. 2], the figure highlights notable differences between the two indicators for some FABs and States, which correlates with the status of FRA implementation (see Figure 4‐17).

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

JAN

FEB

MAR

APR

MAY

JUN

JUL

AUG

SEP

OCT

NOV

DEC

JAN

FEB

MAR

APR

MAY

JUN

JUL

AUG

SEP

OCT

NOV

DEC

JAN

FEB

MAR

APR

MAY

JUN

JUL

AUG

SEP

OCT

NOV

DEC

2012 2013 2014

Flight plan Actual trajectory

4.864.70

3.14

2.722.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

2011

2012

2013

2014


inef

ficie

ncy

(%)


45

Figure 4‐20: Differences flight efficiency by State and FABs (2014)

For those States where FRA is already implemented, the difference between the actual and the planned trajectory is comparatively small with a relatively low level of inefficiency.

At the FAB level, the largest gap of almost 3% between actual and plan, can be observed for FABEC followed by the Blue Med and UK/Ireland FABs, which suggests scope for improvement. As shown in Figure 4‐20, the contribution of States towards inefficiencies within FABs can vary significantly, particularly for the SW and UK‐Ireland FABs.

Route availability and the flexible use of airspace (see the next section) appear to be a contributing factor to the significant differences between the planned and the actual trajectory.

Figure 4‐21 shows the breakdown of the total additional distance in actual trajectories by FAB in 2014. Unchanged from 2013, FABEC accounted for 42% of the total additional distance in 2014, followed by the SW (15%), Blue Med (13%), and UK‐Ireland (12%) FABs.

Figure 4‐21: Share of total actual additional distance by FAB (2014)

Figure 4‐22 shows the average additional distance per flight and the percentage of additional en‐route distance by FAB for the actual trajectories in 2014.

The level of inefficiency is expressed as a percentage and depends not only on the additional distance but also on the average length of the achieved distance.

On average, the additional distance per flight in 2014 was greatest in the SW FAB (17.6km), followed by FABEC, the Blue MED FAB, and the UK‐Ireland FAB.

While the route structure is presently the single most constraining factor, the observed inefficiencies are the result of complex interactions between airspace users, ANSPs and the European Network Manager.

0

1

2

3

4

5

6

7

8

9

UK Continental

Ireland

Spain

Portugal Continen

tal

Switzerland

Belgium

France

Netherlands

Germany

Italy

Cyprus

Greece

Malta

Austria

Czech Rep

ublic

Slovenia

Slovak Rep

ublic

Hungary

Croatia

Lithuania

Poland

Bulgaria

Romania

Estonia

Norw

ay Continental

Latvia

Finland

Swed

enDen

mark

FYROM

Turkey

Albania

Serbia and M

ontenegro

Ukraine

Moldova

Bosnia‐Herzegovina

UK‐IrelandFAB

SW FAB FABEC BLUE MED FAB(Europe)

FAB CE (Europe) BalticFAB

DANUBEFAB

NEFAB DK‐SEFAB

Other

Inefficien

cy (%)

Actual trajectory Flight plan

Source: NM; PRU analysis

(Lines refer to FAB average)

BLUE MED FAB13%

Baltic FAB1%

DANUBE FAB3%

DK‐SE FAB2%FAB CE

8%

FABEC42%

NEFAB1%

OTHER3%

SW FAB15%

UK‐Ireland FAB12%

% of total additional distance (actual trajectory) by FAB in 2014



46

Research is ongoing to better understand and quantify the individual contributing factors (flight planning, awareness of route availability, civil‐military coordination, etc.) in order to identify and formulate strategies for future improvements.

A crucial prerequisite for the development of a better understanding is the collection of better data on the activation of special use airspace and on route availability when the flight plan was submitted by airspace users (shortest route).

Figure 4‐22: Actual additional distance per flight by FAB (2014)

4.4 The effect of the Ukrainian crisis on Network capacity & Network flight efficiency

During 2014, the political crisis in Ukraine affected civil aviation both in Ukraine itself and in neighbouring States. From February onwards the traffic in Ukraine could be seen to decrease compared to 2013, whereas in neighbouring countries it was starting to increase.

Following the downing of MH17 in July 2014, and the closure of airspace in eastern Ukraine, further traffic circumnavigated Ukraine, leading to far above forecasted increases in traffic in Bulgaria, Romania, Hungary, Turkey, and the Slovak Republic (see also section 2.2 ff. on page 4).

Through the tremendous efforts of the staff and management in the relevant ANSPs, this unexpected influx of traffic was handled effectively, with minimal delay to airspace users. The deployment of non‐rostered ATC staff, the re‐organisation of leave and the professionalism and flexibility of controllers meant that the constraints to general air traffic were kept to an absolute minimum.

Figure 4‐23 shows an analysis of the effects of the Ukrainian crisis on horizontal en‐route flight efficiency in Europe. The analysis compares flight efficiency in last filed flight plans in February 2014 and in August 2014.

Figure 4‐23: Impact of the Ukrainian crisis on horizontal flight efficiency

At European level, flight efficiency deteriorated from 4.50% to 4.73% between February and August 2014 (red line).

The vast majority of flights (some 93%) were between city pairs not directly affected by the airspace closure in Ukraine. Flight efficiency for those flights (blue line) improved from 4.85% to 4.73% between February and August.

0%

1%

2%

3%

4%

5%

6%

02468

1012141618

SW FAB

FABEC

BLU

E MED

FAB

UK‐Ireland FAB

FAB CE

DANUBE FA

B

DK‐SE FA

B

NEFAB

Baltic FA

B

inefficien

cy (%)

Additional distance per entry (km)


7%

93%

Flights

February 2014

8%

92%

Extra distance

2.45%

4.77%4.85%

4.73%4.50%

4.73%

February August

Affected by Ukraine crisis Unaffected Europe

Horizontal en‐route flight efficiency (flight plan)


6%

94%

Flights

August 2014

12%

88%

Extra distance


47

The brown line corresponds to the remaining flights, irrespective of their actual use of the Ukrainian airspace. As can be seen in Figure 4‐23, the initially good performance on the affected city pairs deteriorated significantly from 2.45% to 4.77% which is slightly above the European average. The decrease of flight efficiency on those city pairs affected by the Ukrainian crisis offset the observed performance improvement on unaffected city pairs leading to a moderate overall decrease of flight efficiency at European network level.

4.5 Flexible use of Airspace

States have an obligation to meet national security and operational training requirements, as well as meeting the needs of civil airspace users. In order to achieve that, it is occasionally necessary to restrict or segregate airspace for the exclusive use.

To avoid unnecessary constraints in available capacity and flight efficiency, for both civil and military users, airspace restrictions should be based on actual use, and should be cancelled when not required.

As a high‐level indication of the link between the allocation of the areas and their actual use, EUROCONTROL Member States were asked to submit information on the number of allocated hours and the number of hours actually used.

Figure 4‐24 compares the number of hours that airspace was actually used for the activities requiring restriction or segregation and the number of hours that restrictions or segregations were applied for 2014.

Data is not shown for all States in Figure 4‐24, for the following reasons: the requested data was not provided28 (left blank); there were data quality issues (data issues); the States consider that the restriction or segregation of specific areas has no impact either on the available ATC capacity, or on the route options available for general air traffic (not applicable).

State (2014) Used /

Allocated Total hours

State (2014) Used /

AllocatedTotal hours

State (2014) Used /

Allocated Total hours

Albania Germany 45% 33 479 Poland 45% 87 044

Armenia Greece Portugal not appl.

Austria 66% 1 087 Hungary 30% 14 477 Romania 62% 58 362

Belgium Ireland data issues Serbia

Bosnia and Herzegovina. Italy 44% 4 008 Slovakia 57% 55 592

Bulgaria Latvia 29% 635 Slovenia

Croatia not appl. Lithuania 93% 845 Spain 66% 10 321

Cyprus Luxembourg not appl. Sweden 41% 564

Czech Republic 40% 43 098 Malta Switzerland

Denmark 17% 2 655 Moldova FYROM 100% 99

Estonia data issues Monaco not appl. Turkey

Finland 27% 26 262 Montenegro not appl. Ukraine

France Netherlands 84% 28 189 UK 40% 143 634

Georgia not appl. Norway 47% 5 276

Source: States

Figure 4‐24: Ratio of time airspace was used vs. allocated (pre‐tactically)

28 SES States are obliged to provide this information in accordance with Regulation (EU) 691/2010.

Restricted/segregated areas

The map below shows the location of some of the restricted/ segregated areas in the ECAC area.

The map simply shows the geographical locations of published segregated and restricted areas.

The impact assessment of each area depends upon many factors including, but not limited to, position, altitude, period of allocation, demand of General Air Traffic (GAT) at time of allocation, weather, suitability for military operations and training, re‐routing options, distance from base for Operational Air Traffic (OAT) traffic, precise restriction or segregation criteria, etc.


48

Pre‐tactical restrictions were notified the day before operations, in accordance with the Airspace Use Plan (AUP), NOTAM or as published in the national Aeronautical Information Publication (AIP).

Figure 4‐24 gives a high level indication of latent capacity, and flight efficiency opportunities, which could potentially benefit airspace users.

Making the latent capacity and route options available in a predictable manner, when needed by airspace users, will improve the network planning of available capacity and flight efficiency to meet the airspace users requirements, thus providing a better air navigation service.

With a number of States showing the airspace is actually used less than 50% of the time that it is reserved for exclusive use, there is clearly scope for improvement.

Review of civil military cooperation and coordination arrangements.

Further to the work in PRR 2013, the PRC contacted three member states and asked them to provide voluntary information on how civil military stakeholders cooperate and coordinate to provide the optimum benefit for airspace users.

Following a review of the FUA concept, and Articles 6 – 9 of the FUA Regulation (EC No 2150/2005) [Ref. 18], the PRC has identified several criteria that it considers as essential to effective cooperation and coordination between civil and military stakeholders.

CRITERIA Key Findings

a

Identification of specific airspace that, when restricted or segregated, can affect the availability of route options or the availability of ATC capacity for general air traffic

Although national AIPs list all potential restricted and segregated airspace, no distinction is made between those areas that have an impact on ATC capacity and available route options, and those that do not. Airspace users have limited transparency on how the airspace structures are managed.

Without a clear identification of the specific airspace that impact operation of GAT, the effectiveness of civil military cooperation and coordination is limited.

b

Assessment of the impact of each area listed in (a.) on the availability of route options or on the reduction of ATC capacity for general air traffic;

Airspace users have no information regarding impact assessments relating to the restriction or segregation of airspace that affects their operations.

In the 3 Member States there are 3 different approaches:

Generic 10% reduction in available ATC capacity when there is a restriction or segregation of airspace;

Permanent setting of a reduced ATC capacity level which presumes restriction or segregation of airspace;

Evaluation of impact according to internal procedures, the results of which are not necessarily communicated to NM or airspace users.

c

Agreement between (Level 1) civil and military stakeholders of the strategic objectives to be accomplished within the State/region;

The 3 Member States could provide no evidence of Strategic Objectives that have been agreed between civil and military stakeholders.

d

Promulgation of priorities and procedures for the management of national / regional airspace in accordance with the strategic objectives stated in (c.) and feedback from (i) below;

Although one Member State has published a document detailing priorities and procedures for Airspace Management, the absence of level 1 Strategic Objectives means that it is impossible to benchmark the effectiveness of those procedures and priorities.

Neither of the other Member States has promulgated priorities and procedures for the management of airspace.

Without Strategic Objectives it is impossible to determine whether or not feedback from levels 2 and 3 has been considered by level 1.


49

e

Exchange of pre‐tactical demands, constraints and opportunities for civil and military airspace requirements;

2 States have an exchange of pre‐tactical demands, constraints and opportunities for civil and military airspace requirements.

The third State only exchanges military airspace demands.

f

Discussion, based on (a) to (e) above, between civil and military airspace managers, with the objective and empowerment to achieve the optimum allocation of airspace in accordance with the national/regional procedures and published strategic objectives;

Although discussions between civil and military airspace managers (level 2) take place, the absence of national/regional procedures and published Strategic Objectives make it impossible to determine whether or not the airspace has been allocated for the optimum benefit of all airspace users.

g Notifying airspace users of the allocation/restriction/availability of airspace;

All 3 Member States use the pre‐tactical AUP process to notify airspace users of airspace restrictions.

However, the 3 States follow different approaches in regards to the tactical notification of the airspace situation on the day of operations:

One State notifies NM, and thence airspace users, of all changes to airspace availability that impacts GAT;

The second State applies an internal filtering process and only notifies NM of ‘significant’ changes in the airspace situation;

The third State does not notify NM and the airspace users of changes in the airspace situation.

h

Post‐Operations Monitoring of the impact of the airspace management decisions on both general and operational air traffic

The lack of Strategic Objectives prevents effective assessment of the airspace management decisions.

The 3 Members States are only able to assess the impact of airspace management decisions in terms of delay and flight inefficiency for GAT.

Since there are no published performance objectives for OAT it is impossible to assess the impact of airspace management decisions on OAT.

i

Regular review of (h) and feedback to Level 1 stakeholders: including problems, issues and requests for change to the priorities and procedures listed in (d) above.

For all 3 Member States:

Due to the lack of performance objectives for military airspace users, feedback to level 1 stakeholders is only possible in regards to capacity and route availability for GAT.

Since the level 1 civil‐military stakeholders have not agreed national Strategic Objectives it is impossible to derive the effectiveness of the review and feedback procedure.

Figure 4‐25: Criteria to effective cooperation/ coordination between civil and military stakeholders

The PRC gathered information on each of the above criteria during interviews with airspace managers in each of the three member states. Since the purpose of this research was to gather information on the variance of the application of FUA, rather than identify compliance or lack of compliance with the legislation, the PRC will only highlight the differences, rather than specifying which State does what.

A review of 3 Member States identified 3 very different approaches to civil military cooperation and coordination, despite the existence of a EUROCONTROL FUA specification and binding EU legislation on the Flexible Use of Airspace (since 2005).

The PRC believes that collecting data related to arrangements on civil military cooperation and coordination in each of the EUROCONTROL Member States, based on the nine criteria listed above, will identify areas where improvements can be made and enable the sharing of best practices which will realise benefits for both civil and military airspace users. Such a collection of data could be useful in updating the FUA specification.

The PRC is extremely grateful to the three Member States involved, and to the personnel who kindly provided their time and efforts in assisting the PRC with its analysis.


50

4.6 European ATFM performance (network level)

Over the past few years, the role of the network function in Europe has been strengthened by the SES legislation. This evolution foresees a more proactive role in Air Traffic Flow Management (ATFM), ATC capacity enhancement, route development, and the support to the deployment of technological improvements across the ATM network for the Network Manager.

EUROCONTROL has been entrusted by the European Commission with the SES Network functions. These functions are laid out in Commission Regulation (EC) N°677/2011 [Ref. 19] and include, inter alia, the tasks foreseen for the Central Unit for Air Traffic Flow Management (ATFM).

The ATFM function in Europe is jointly executed by local ATFM units and the Network Manager (central unit for ATFM). ATFM regulations are put in place by the Network Manager to protect en‐route sectors or airport from receiving more traffic than ATC can safely handle upon request of the local Flow Management Positions (FMP).

Figure 4‐26 shows the evolution of the three high‐level indicators presently in use to monitor the performance of the ATFM function at system level.

A number of initiatives have been promoting ATFM slot adherence in order to improve traffic predictability [Ref. 20]. Local ATC at the respective departure airport has a joint responsibility with aircraft operators to make sure that the aircraft depart within the allocated ATFM window in order to avoid over‐deliveries which occur when more aircraft than planned enter a protected sector (see also ATFM slot adherence at airports in Chapter 5).

The corresponding indicator in Figure 4‐26 shows that the share of take‐offs outside the ATFM slot tolerance window (‐5min +10 min) continuously improved between 2003 and 2014.

The share of regulated hours with over deliveries in Europe remained more or less constant around 11% between 2011 and 2014. Future efforts should aim at reducing the level as much as possible to further increase system confidence which can in turn free latent capacity kept as a reserve to protect controllers from excessive workload.

The share of avoidable ATFM regulations (i.e. there was not excess demand) increased slightly between 2011 and 2014. The ability for improvement is largely linked to predictability and accuracy of the relevant information when the decision to call for an ATFM regulation is taken (i.e. several hours before the anticipated capacity shortfall).

Enhanced traffic projections through A‐CDM implementation at more airports (see also Chapter 5), enhanced data quality but also improvements in aviation metrological capabilities could help improving performance in this area.

ATFM performance assessment

Regulation (EC) No 255/2010 [Ref. 23] of 25 March 2010 laying down common rules on air traffic flow management aims at optimising the available capacity of the European air traffic management network (EATMN) and enhance air traffic flow management (ATFM) processes by establishing requirements for ATFM.

It requires, inter alia, the central unit for ATFM to produce annual reports indicating the quality of the ATFM in the airspace of the Regulation including causes of ATFM measures, impact of measures and adherence to ATFM measures.

Figure 4‐26: ATFM performance (network indicators)

0.0%

2.5%

5.0%

7.5%

10.0%

12.5%

15.0%

17.5%

20.0%

22.5%

25.0%20

03

200

4

200

5

200

6

200

7

200

8

200

9

201

0

201

1

201

2

201

3

201

4% of take offs outside ATFM slot tolerance window

% regulated hrs. with actual demand/capacity >110%(excess demand)

% of ATFM delays due to avoidable regulations (noexcess demand)

Source: Network Manage


51

4.7 Conclusions

After the continuous improvement between 2010 and 2013, (reaching the lowest level of en‐route ATFM delay per flight on record in 2013 (0.53 minutes per flight)), en‐route ATFM delays in the EUROCONTROL area increased again to 0.61 minutes per flight in 2014.

The performance deterioration in 2014 was mainly attributed to ATC capacity issues confirming concerns already highlighted in PRR 2013 that ATFM delays could spike, when traffic grows again, unless sufficient attention is focussed on capacity planning and deployment.

In 2014, European traffic flows and capacity management were affected by a number of significant events including the downing of Malaysian Airline MH17 in Ukrainian airspace in July. The ensuing total closure of airspace in the eastern part of Ukraine generated many re‐routings in that region. Traffic in Bulgaria, Romania, Hungary, Turkey, and Slovakia increased far beyond forecast levels but, thanks to the efforts of those ANSPs, they were accommodated without significant delays.

While capacity constraints can occur from time to time, area control centres (ACCs) should not generate high delays on a regular basis. In 2014, the most constraining ACCs were Nicosia (failure to implement capacity plans), Warsaw (issues with the implementation of the new PEGASUS 21 system), Lisbon (reoccurring issues in November), Canarias (capacity planning and delay classification issues), Athinai/Macedonia (insufficient capacity in summer), Reims (capacity planning), Brest and Marseille (ATC industrial action) which accounted together for more than half of all European en‐route ATFM delays (54.6% ) but only for 17.8% of total flight hours controlled.

It is essential, particularly in view of the considerable lead times, to carefully plan and also deploy capacity in line with projected traffic growth. Over‐conservative capacity planning removes buffers against unexpected traffic increases and increases the risk of significant disruption to aircraft operations.

Although the unexpected changes in traffic flows in 2014 had a moderate negative effect on European flight efficiency in 2014, overall, there was a notable improvement in 2014, confirming the positive trend observed over the past years. Coordinated by the Network Manager, 30 of the 64 ACCs had implemented various steps of Free Route Operations by the end of 2014 with clear benefits in terms of flight efficiency.

In 2014, the level of inefficiency in filed flight plans in Europe decreased from 4.86% to 4.70% and actual trajectories improved at an even higher rate from 3.14% to 2.72%. Whereas the stronger improvement in actual trajectories is positive in terms of fuel burn and CO2 emissions, the widening of the gap between planned and actual flown trajectories needs to be monitored closely as it has a negative impact on network predictability.

A voluntary review of only three Member States identified significant differences, not only between the national arrangements and the FUA specification but also between the individual Member States, in how the airspace is managed to provide the optimum benefit for civil and military airspace users. A wider review could help to better identify and understand existing shortcomings and identify best practice for the future benefit of all airspace users.

Chapter 5: Operational ANS Performance at Airports

PRR 2014 ‐ Chapter 5: Operational ANS Performance ‐ Airports 52

5 Operational ANS Performance at Airports


On average, IFR movements (arrival + departure) at the top 30 airports increased by 2.7% in 2014.

In 2014, the average additional time for ASMA and taxi‐out showed a slight improvement in 2014. Airport ATFM arrival delays and ATC pre‐departure delays increased at the same time.

The underlying reasons for capacity/demand imbalances differ by airport and range from the airport scheduling process to the ability to sustain declared capacity in various circumstances on the day of operations.

Data exchange (A‐CDM) and increased situation awareness is a key enabler for performance improvements. However the mere availability of information does not automatically resolve issues without proactively addressing relevant optimisation processes.

European average (top 30 airports)

2014 change vs.

2013

Average daily movements (dep.+ arr.)

21 353 +2.7%

Avg. Airport Arrival ATFM Delay (min./arr.)

0.9 +0.1 min/arr.

Avg. Additional ASMATime (min./arr.)

2.1 ‐0.1

min./arr.

Avg. ATC Pre‐departure Delay (min./dep.)

0.9 +0.1

min./dep.

Avg. Additional Taxi‐out Time (min./dep.)

3.5 ‐0.2

min./dep.

5.1 Introduction

Airports are key nodes in the air transport network and the availability of sufficient capacity is frequently identified as being one of the main bottlenecks for future growth of the European air transport system.

According to the latest Challenges of growth 2013 report [Ref 21], the economic downturn starting in 2008 led to a significant downward adjustment of airport capacity expansion plans from 38% to 17% over the next 20 years. In view of the considerable downward adjustment of expansion plans in Europe, the optimised use of available airport capacity is crucial to keep delays to a minimum.

ANS plays a key role in balancing traffic with available capacity at airports within the given infrastructural and environmental constraints and in the integration of airports in the European network. This chapter provides a review of operational ANS performance at major European airports. It is to be noted that the evaluation of future airport capacity requirements (e.g. new runway, taxiways, etc.) is beyond the scope of this report.

Although the focus of this chapter is on ANS performance at airports, it is important to point out that performance is the result of complex interactions between stakeholders (airlines, ground handlers, airport operator, ATC, slot coordinator, etc.) and a clear identification of underlying causes can be difficult. While at airports, ANS is often not the root cause for a capacity/demand imbalance (e.g. policy decisions in the airport scheduling phase, traffic demand variation, weather disruptions) the way traffic is managed and distributed along the various phases (air vs. ground) has a different impact on airspace users (time, fuel burn, costs), the utilisation of capacity, and the environment (emissions).

Consequently, the analyses in the respective sections of this chapter should not be interpreted in isolation, but as an integral part of the overall operational performance observed at the corresponding airport.

Due to the large number of factors affecting performance (e.g. weather), the desired (e.g. environmental) or required (e.g. safety) constraints and the level of flexibility at which the ATM system is operated today (i.e. 15 minute tolerance windows), a certain level of delay can be necessary or may even be desirable, if a system is to be run efficiently without underutilisation of available resources. In view of these operational trade‐offs and exogenous factors, it is not possible to reduce inefficiencies to zero while maximising the use of capacity at the given level of flexibility.

PRR 2014 ‐ Chapter 5: Operational ANS Performance ‐ Airports

53

The analysis in this chapter focuses on the top 30 airports in terms of IFR movements in 2014 as they have the strongest impact on network‐wide performance. Together they accounted for 45.5% of all European departures in 2014.

As part of the regular operational ANS performance review at European airports, more than the 30 airports included in this chapter are continuously monitored. Any unusual performance observed at an airport not included in the top 30 will be commented on in the respective sections of the chapter. Information on the observed performance at all airports is summarised in Annex VII of this report.

The data used in this chapter are based on airport operators’ data submissions in accordance with the established technical specification for the airport operator data flow, complemented by data from the Network Manager. The implementation of the specification is on‐going and for some of the top 30 airports the data are not available at the required level of quality or granularity (e.g. stand or runway information). If an indicator could not be computed because of data availability issues, no results are displayed for the respective airports in this chapter.

The technical specification for the airport operator data flow ensures a harmonised approach to ANS performance review as it encapsulates the requirements of the SES performance scheme and provides uniform data standards for all States in the EUROCONTROL area.

5.2 Traffic evolution at the main European airports

Figure 5‐1 shows the evolution in average daily movements at the top 30 airports in absolute and relative terms. On average, IFR movements (arrival + departure) at the top 30 airports increased by 2.7% in 2014 compared to 2013 but are still ‐3.3% below 2008 levels. Compared to 2013, the number of passengers at the top 30 airports increased by +5.8% in 2014 (9.9% above 2008 levels).

Figure 5‐1: Traffic variation at the top 30 European airports (2014/2013)

Paris Charles de Gaulle, Munich, Paris Orly and Frankfurt airport showed the highest decrease in absolute terms in 2014. As a result of the traffic decrease at Paris (CDG) and Frankfurt, London Heathrow surpassed the two airports and now ranks first in terms of movements in Europe.

The two Turkish airports, Istanbul Ataturk and Sabiha Gökçen, continued their remarkable growth already observed in previous years. The increase in average daily traffic was close to 100 additional flights at both airports in 2014. In relative terms, Istanbul’s Sabiha Gökçen was the fastest growing airport with an impressive +23.9% traffic increase year on year.


54

5.3 Cancellations at European airports

This section provides an analysis of the level of operational cancellations at European airports. Figure 5‐3 shows the share of operational cancellations at 24 of the 30 airports29 in 2014 which have started reporting regularly on the number of cancellations.

Although there were peaks up to 12% cancellations on some days due to special events, a continuous level of cancellations can be observed. Overall, the average cancellation rate in 2014 was 1.5% of the planned movements.

Operational cancellations at airports In accordance with Regulation (EC) 691/2010, a flight is considered to be cancelled if the following conditions apply:

The flight received an airport slot;

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 operations; but,

The actual landing or take –off never occurred.

Figure 5‐2: Average daily operational cancellations in 2014 (24 airports included)

The observed continuous level of cancellations may be linked to some extent to the current airport slot coordination mechanism in Europe. In order to keep the same series of airport slots in the following season, air transport operators are required to use a series of slots at least 80% of time during the season (“grandfather rights”).

Locally, there were notable variations by airport. The highest level of operational cancellations (4.3%) was observed at Geneva airport, followed by Lisbon, and Vienna (see Figure 5‐3). More analysis is needed to better understand the underlying drivers at each of the airports.

Figure 5‐3: Operational cancellations at the top 30 airports in 2014

29 No data is presently available for the Paris (CDG), Istanbul (IST), Paris (ORY), Dusseldorf (DUS), Antalya (AYT), and

Istanbul (SAW).

0%

2%

4%

6%

8%

10%

12%

01/01/2014

01/02/2014

01/03/2014

01/04/2014

01/05/2014

01/06/2014

01/07/2014

01/08/2014

01/09/2014

01/10/2014

01/11/2014

01/12/2014

Source: PRU analysis; CODA

Average daily cancellation rate (24 of top 30 European airports)

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

4.5%

London* (LHR)

Paris* (CDG)

Frankfurt* (FRA)

Amsterdam

(AMS)

Istanbul A

taturk (IST)

Munich* (M

UC)

Madrid* (M

AD)

Rome*

(FCO)

Barcelona (BCN)

London* (LGW)

Zurich* (ZRH)

Copen

hagen

(CPH)

Oslo* (OSL)

Vienna (VIE)

Paris Orly (ORY)

Stockholm

(ARN)

Brussels* (BRU)

Duesseldorf* (DUS)

Gen

eva (GVA)

Berlin Tegel (TXL)

Istanbul (SA

W)

Dublin

(DUB)

Antalya (AYT)

Palma (PMI)

Manchester (M

AN)

Helsinki* (HEL)

Milan* (M

XP)

Lisbon (LIS)

London (STN)

Athens (ATH

)

% of operational cancellations by airport (2014)

no data

* A‐CDM implemented airport

Source: PRU analysis; CODA

no data

no data

no data

no data

no data

Average (2014)


55

5.4 Balancing capacity with demand at airports

With rare exceptions such as Berlin Tegel where the terminal capacity is the limiting factor, the number of operations at airports is usually limited by the runway system capacity. In addition to physical constraints, such as runway layout, there are “strategic” factors such as airport scheduling and “tactical” factors which include, inter alia, the sequencing of aircraft and the sustainability of throughput in specific weather conditions.

While safe operation of aircraft on the runway and in the surrounding airspace is the dominant factor influencing runway throughput, other influencing factors comprise: airport layout and runway configuration, traffic mix, runway occupancy time of aircraft during take‐off and landing, separation minima, wake vortex, ATC procedures, weather conditions, demand loading by departure route, and environmental restrictions.

Meteorological conditions can have a major impact on the operational capacity of an airport. As weather conditions deteriorate, separation requirements generally increase and the actual throughput is consequently reduced.

The impact of weather (visibility, wind, convective weather, precipitation, etc.) on operations at an airport and hence on ANS performance can vary significantly by airport and depends on a number of factors such as, inter alia, ATM and airport equipment (instrument approach system, radar, de‐icing facilities, etc.), runway configurations (wind conditions), and approved rules and procedures.

As mentioned above, airports in Europe are designated as ‘coordinated’ when the airport capacity is insufficient to fulfil airlines’ demand during peak hours. In order to avoid frequent and significant excess of demand on the day of operations, airport capacity declaration and the subsequent airport scheduling process regulate traffic in terms of volume and concentration by allocating airport landing and departure slots to aircraft operators months before the actual day of operation. Of the top 30 airports evaluated in this chapter, all but Athens are coordinated (IATA Level 3)30. Taking into account the level of congestion, further research is required to analyse the benefits of the introduction of the coordination process at – currently – uncoordinated airports.

The declared airport capacity31 is generally decided by the respective States taking into consideration the opinion of a coordination committee32 reflecting also infrastructure limitations and environmental constraints. It represents an agreed compromise between the maximisation of airport infrastructure utilisation and the quality of service considered as locally acceptable. This trade‐off is usually agreed between the airport managing body, the airlines and the local ATC provider during the airport capacity declaration process. A declared airport capacity close to Instrument Meteorological Conditions (IMC) can support overall stability of operations but there is a risk that resources might be underutilised for considerable periods.

The recent US EUR study [Ref. 22] shows that average throughput at comparable airports is higher in the US than in Europe. In these cases, the capacity declaration is closer to Visual Meteorological Conditions. This approach can result in high instability of operations during adverse weather periods which impede the application of VMC operations.

Depending on the economic value of the airport slot for aircraft operators, a higher level of “planned” delay is accepted by airlines at some airports as a trade‐off to get access to the airport

When a mismatch between demand and capacity is anticipated at the airport on the day of operations, a number of air traffic flow management techniques can be applied on the arrival and departure traffic flow, depending on the anticipated duration and severity of the capacity shortfall.

The following sections evaluate ANS‐related inefficiencies on the departure and arrival traffic flow at

30 See latest edition of the IATA World Scheduling Guidelines and Regulation (EC) 793/2004.

31 The airport capacity declaration is a local process and can vary by airport. There is no harmonised method to

declare an airport’s capacity in Europe. 32 The responsibility to set up a coordination committee lies with the respective State.


56

the top 30 airports. The four performance indicators used for the analysis are illustrated in dark blue in Figure 5‐4. They relate to the optimisation of the inbound and outbound traffic flow and are also part of the SES performance scheme. Complementary to the four standard indicators, an analysis of taxi‐in efficiency is provided.

Figure 5‐4: ANS‐related efficiency at airports

There is a close interrelation between the management of the inbound and outbound traffic flow and the efficiency and predictability of the turn‐round phase at airports. The Airport Collaborative Decision Making (A‐CDM) concept was developed to facilitate the sharing of operational processes and data to allow better informed decisions to be made at local but also at network level. One of the fundamentals of the CDM Milestone Approach is the real‐time sharing of milestones such as Target Off‐Block Time, thus creating “common situation awareness” among involved stakeholders and improved utilisation of resources.

Locally, the A‐CDM concept aims at improving the overall efficiency of operations at the airport by synchronising activities of all involved players to maximise the departure predictability and optimisation of resources, with a particular focus on the aircraft turn‐round and pre‐departure sequence. The turn‐round phase consists of a number of complex and interrelated pre‐departure processes (re‐fuelling, catering, baggage, boarding, etc.) with involvement from different stakeholders.

Figure 5‐5: Average inbound and outbound delays at the top 30 airports in 2014

Delays already encountered at the departure airport are by far the main contributor towards arrival punctuality at the destination (see also Chapter 2). Figure 5‐5 provides a first insight on the efficiency of the turn‐round phase by comparing the average arrival and departure delay at the top 30 airports in 2014. Turn‐round performance if affected by many factors including airline scheduling (i.e time buffers in schedules). In view of the considerable costs involved, many airports have local working groups dedicated to improving the turn‐round efficiency.

Rome Fiumicino, Paris Orly, London Gatwick and Lisbon have the most notable delay amplifying effect between inbound and outbound traffic flow. It is interesting to note that Rome Fiumicino and London Gatwick are fully implemented A‐CDM airports. On average, the difference between A‐CDM

‐5

0

5

10

15

20

London* (LHR)

Paris* (CDG)

Frankfurt* (FRA)

Amsterdam

(AMS)

Istanbul A

taturk (IST)

Munich* (MUC)

Madrid* (M

AD)

Rome*

(FCO)

Barcelona (BCN)

London* (LGW)

Zurich* (ZRH)

Copen

hagen

(CPH)

Oslo* (OSL)

Vienna (VIE)

Paris Orly (ORY)

Stockholm

(ARN)

Brussels* (BRU)

Duesseldorf* (DUS)

Gen

eva (GVA)

Berlin Tegel (TXL)

Istanbul (SA

W)

Dublin

(DUB)

Antalya (AYT)

Palma (PMI)

Manchester (M

AN)

Helsinki* (HEL)

Milan* (M

XP)

Lisbon (LIS)

London (STN)

Athen

s (ATH

)

minutes

Avg. arrivaldelay

Avg. departuredelay

Differencedeparture vs.arrival delay

Source: CODA ; PRC analysis* A‐CDM implemented airport

ANS‐related efficiency at airports

Turn around phase

(efficiency & predictability)

Arrival flow management Departure flow management

Taxi‐in

Approach (ASMA)

Airport arrival ATFM delay

ATC‐related departure delay

Taxi‐out additional

time

Optimisation of departure sequencing

Balancing ATFM delays at origin airport vs. local airborne holdings


57

airports (delta of 1.7 minutes between inbound and outbound flow) and non A‐CDM airports (1.8 min. per flight) is not as significant as would be expected as a result of the performance at Rome Fiumicino and London Gatwick. This stresses the point that the main objective of A‐CDM is focused on predictability and not on reducing the inbound and outbound flow delays.

To date, A‐CDM has been fully implemented at 15 European airports, which corresponds to 24.1% of all European departures in 2014. In the analysis in the next sections, the fully implemented A‐CDM airports are marked with an asterisk.

With A‐CDM implemented locally at an airport, the next steps are to enhance the integration of airports with the Network Manager (NM) for the benefit of the entire network. Exchange of real time data between airports and the NM is already operational at the 15 fully implemented A‐CDM airports. The airports are receiving a more accurate arrival estimate for all flights via the Flight Update Message (FUM) and the network is benefiting with enhanced take‐off time estimates in tactical operations via the Departure Planning Information (DPI) messages.

This better integration of airports into the ATFM network in turn leads to a higher accuracy of the traffic situation and hence a better utilisation of the entire network capacity. For 2015, a further 8 airports are foreseen to become fully A‐CDM implemented.

5.5 Management of the arrival flow

Apart from the capacity declaration process in the strategic phase, the primary means to manage arrival flows at airports in Europe in the tactical phase are ATFM regulations centrally implemented by the Network Manager and local flow measures at the airport.

Small imbalances between demand and capacity during peak times are usually managed locally by holdings or vectoring which may also serve as a short‐term buffer to ensure a constant reservoir of aircraft to maximise runway throughput. In case of a more severe imbalance when delays cannot be absorbed around the airport, the Flow Management Position (FMP) coordinates with the Network Manager the application of an ATFM regulation which will hold aircraft bound for the capacity constrained airport at their origin airports.

The level of accuracy of the flow measures significantly increases from the application of ATFM departure slots at the origin airports (15 min. time window) to holding or vectoring in the vicinity of the arrival airport.

In the absence of a supporting en‐route function in Europe, finding the right balance for the management of the arrival flow can be challenging. Although keeping an aircraft at the gate saves fuel – if it is held and capacity goes unused – the cost to the airlines of the extra delay may exceed the fuel cost by far.

Declared arrival capacity vs. actual throughput

This section compares the declared arrival capacity to actual throughput at the top 30 European airports. It provides an indication of the arrival throughput “peak service rate” that can be achieved in ideal conditions and with a sufficient supply of demand. It is important to point out that the peak service rate cannot necessarily be sustained when conditions are not ideal and it should therefore not be mistaken as an indication that the declared capacity can be increased.

Airport peak service rate

The peak service rate (or peak throughput) is an approximation of the operational airport capacity that was provided in ideal conditions. It is based on the cumulative distribution of the number of movements per hour, on a rolling basis of 5 minutes.

The comparison also shows those airports with spare capacity (i.e. peak service rate is notably below peak declared capacity).

Figure 5‐6 shows the declared peak arrival capacity in 2014 together with the computed “peak service rate” which gives an indication of the actual peak throughput at the airport. Complementary

Arrival flow management

Taxi‐in

Approach (ASMA)


Balancing ATFM delays at origin airport vs. local airborne holdings


58

to the 2014 results, the peak service rate before the start of the economic crisis in 2008 is shown.

The Istanbul Ataturk airport shows a substantial increase in peak arrival throughput compared to 2008 which is consistent with the sustained traffic growth observed in Turkey.

Following the opening of the new north‐west runway at the end of 2011, the peak arrival service rate increased notably at Frankfurt airport compared to 2008 and is now well above the peak declared arrival capacity. The new runway is only used for arrivals and enables the airport to handle independent parallel landings.

Milan Malpensa, Madrid and Rome Fiumicino show the most significant decline in peak arrival throughput compared to 2008.

With the exception of Frankfurt, Paris Charles de Gaulle, London Heathrow, Zurich, and Palma, the peak service rate was below the declared peak arrival capacity.

Figure 5‐6: Peak declared arrival capacity and service rate

ANS‐related inefficiencies on the arrival flow

This section analyses ANS‐related inefficiencies on the arrival flow at the top 30 European airports in terms of airport arrival ATFM delay and additional ASMA time. Although the direct influence is limited, for completeness reasons and to understand the order of magnitude, the efficiency of the taxi‐in performance is also included.

ANS in Europe are still largely organised by state boundaries. This means that national service providers are usually able to directly influence an aircraft only once it enters its airspace, which is sometimes close to the destination airport. This limits the opportunity to manage the approach already during the en‐route or descent phase through the application of speed control and can result in additional time in holding stacks.

Additional ASMA Time

ASMA (Arrival Sequencing and Metering Area) is the airspace within a radius of 40NM around an airport. The Additional ASMA Time is a proxy for the level of inefficiency (holding, sequencing) of the inbound traffic flow during times when the airport is congested.

The methodology is described in more detail online [Ref. 15].

In an initiative to already absorb some of the stack holding times in the more fuel efficient en‐route phase, the cross border arrival management (XMAN) project was set up for major arrival flows into London Heathrow airport in March 2014. The neighbouring ANSPs (DSNA, IAA, MUAC, LVNL) were asked to slow down aircraft up to 350 miles away from London to help minimise local holding delays at London Heathrow by two minutes by the end of 2014.

44 62 55 68 29 58 48 54 38 35 36 52 34 48 34 42 48 33 22 30 29 33 33 48 40 26 31 3515

25

35

45

55

65

London* (LHR)

Paris* (CDG)

Frankfurt* (FRA)

Amsterdam

(AMS)

Istanbul A

taturk (IST)

Munich* (M

UC)

Madrid* (M

AD)

Rome*

(FCO)

Barcelona (BCN)

London* (LGW)

Zurich* (ZRH)

Copen

hagen

(CPH)

Oslo* (OSL)

Vienna (VIE)

Paris Orly (ORY)

Stockholm

(ARN)

Brussels* (BRU)

Duesseldorf* (DUS)

Gen

eva (GVA)

Berlin Tegel (TXL)

Istanbul (SA

W)

Dublin

(DUB)

Antalya (AYT)

Palma (PMI)

Manchester (M

AN)

Helsinki* (HEL)

Milan* (MXP)

Lisbon (LIS)

London (STN)

Athen

s (ATH

)

Arrivals per hour

Declared arrival capacity (2014)

2008 (actual)

2014 (actual)

Peak declared arrival capacity and service rate

Source: PRU analysis* A‐CDM implemented airport


59

Figure 5‐7 shows the airport arrival ATFM delay (top of figure) and the additional ASMA time (bottom of figure) per arrival at the top 30 airports.

Together, the top 30 airports accounted for 86% of all airport ATFM arrival delay in Europe. As could also be observed in previous years, weather (66%) and ATC capacity and staffing (28%) were by far the main reason for ATFM airport arrival delays in 2014.

Figure 5‐7: ANS‐related inefficiencies on the arrival flow at the top 30 airports in 2014

On average, ASMA additional time is more than twice (2.1 minutes per arrival) the level of ATFM airport arrival delay (0.9 min. per arrival) in 2014, which is largely driven by the high level of ASMA additional time at London Heathrow.

Following the continuous improvement between 2010 and 2013, there has been a slight increase in average airport ATFM arrival delay in 2014, mainly due to capacity‐related issues.

At airport level, Zurich, Istanbul Sabiha Gökçen and London Heathrow showed the highest average delay per arrival in 2014. Whereas delays at Heathrow were almost all weather‐related, a notable share of the arrival ATFM delays at Zurich and Sabiha Gökçen were capacity‐related.

Average additional ASMA time in Europe continued to reduce in 2014. The improvement was largely driven by the reduction at London Heathrow with a positive effect from the cross‐border arrival management trial which started in March 2014. Despite this improvement, London Heathrow had again by far the highest level of ASMA additional time per arrival in Europe. It should however be noted that the high level of ASMA additional time at London Heathrow is not related to poor ATM performance but due to a deliberate decision taken during the airport scheduling process after consultation with involved stakeholders.

In addition to the flight perspective which evaluates performance in relative terms (average per flight), Figure 5‐8 shows the airports’ contribution towards European network performance which combines the performance with the traffic volume (share of total delay minutes per ASMA additional time and airport arrival ATFM delay).

London Heathrow had by far the highest impact on the European network with 29.3% of total ASMA additional time and 13.2% of total airport arrival ATFM delays in 2014.

012345678910

London* (LHR)

Paris* (CDG)

Frankfurt* (FRA)

Amsterdam

(AMS)

Istanbul A

taturk (IST)

Munich* (M

UC)

Madrid* (M

AD)

Rome* (FCO)

Barcelona (BCN)

London* (LGW)

Zurich* (ZRH)

Copenhagen (CPH)

Oslo* (OSL)

Vienna (VIE)

Paris Orly (ORY)

Stockholm

(ARN)

Brussels* (BRU)

Duesseldorf* (DUS)

Gen

eva (GVA)

Berlin

Tegel (TXL)

Istanbul (SA

W)

Dublin

(DUB)

Antalya (AYT)

Palma (PMI)

Manchester (MAN)

Helsinki* (HEL)

Milan* (M

XP)

Lisbon (LIS)

London (STN)

Athens (ATH

)

min. per arrival

2014

2013

0

1

2

3

4

5

6

7

8

9

10

Other (allother codes)

Weather (C,D)

ATC other(codes IRVT)

Cap./ staffing(codes CSG)

2013

Airport ATFM arrival delays

0.0

0.5

1.0

1.5

2.0

2.5

2010

2011

2012

2013

2014

0.0

0.5

1.0

1.5

2.0

2.5

2010

2011

2012

2013

2014

ASMA additional time


not available

not available

not available

not available



60

Figure 5‐8: Airport contribution towards European network performance (arrival flow)

Figure 5‐9 shows the average daily traffic profile (left side) and the breakdown of weather‐related airport arrival ATFM delays.

Figure 5‐9: Performance drivers at London Heathrow airport

As can be seen on the left side of Figure 5‐9, due to the high economic value of an airport slot at London Heathrow, the schedule intensity is very high with continuous arrivals and take‐offs (close to declared capacity) throughout the day making the airport one of the busiest two‐runway airports worldwide. In order to give the communities at the end of each runway a predictable respite from noise, the airport normally operates in ‘segregated mode’, with one used for arrival and the other for departures. Although ‘mixed mode’ operations, when both runways are used for both arrivals and departures, would provide additional capacity this is presently not considered as a sustainable option because of the increased noise exposure.

Constant operations close to maximum capacity at capacity‐constrained airports susceptible to adverse weather may potentially result in major delays and cancellations when there is a mismatch between scheduled demand and available capacity.

As can be seen on the right side of Figure 5‐9, wind is by far the most dominant weather phenomenon affecting the airport’s acceptance rate and resulting in high levels of ATFM arrival delays. Today, flights are separated by set distances dependent on the wake vortex category. Strong headwind conditions result in slower aircraft ground speeds and thus extra time between each arrival if distance‐based separation is maintained. The resulting reduced landing rate can lead to substantial delays at capacity‐constrained airports.

A new system to separate arriving aircraft at London Heathrow by time instead of distance set to become operational in spring 2015 is expected to significantly reduce ATFM delays due to high headwinds at the airport.

0%

5%

10%

15%

20%

25%

30%

35%

London* (LH

R)

Paris* (CDG)

Frankfurt* (FRA)

Amsterdam

(AMS)

Istanbul A

taturk (IST)

Munich* (MUC)

Madrid* (MAD)

Rome*

(FCO)

Barcelona (BCN)

London* (LG

W)

Zurich* (ZRH)

Copen

hagen

(CPH)

Oslo* (OSL)

Vienna (VIE)

Paris Orly (ORY)

Stockholm

(ARN)

Brussels* (BRU)

Duesseldorf* (DUS)

Gen

eva (GVA)

Berlin

Tegel (TXL)

Istanbul (SA

W)

Dublin

(DUB)

Antalya (AYT)

Palma (PMI)

Manchester (M

AN)

Helsinki* (HEL)

Milan* (M

XP)

Lisbon (LIS)

London (STN)

Athens (ATH

)

% of total (top 30 airports)

Airport contribution towards total network performance (arrival flow)

ASMA additional time



0

5

10

15

20

25

30

35

40

45

50

0600

0645

0730

0815

0900

0945

1030

1115

1200

1245

1330

1415

1500

1545

1630

1715

1800

1845

1930

2015

2100

2145

2230

Local time

Hourly arrival rate at London (LHR) airport in 2014


51.5%64.8%

57.5%

0%

20%

40%

60%

80%

100%

2012 2013 2014% of weather relatede ATFM

arrival

delay

Not specified

Precipitation

TS/CB

Visibility

Wind


Breakdown of weather‐related airport ATFM arrival delays at London (LHR)


61

With a considerable lower traffic volume than London Heathrow, Zurich airport accounted for 10.3% of total airport arrival ATFM delays and 6.0% of all ASMA additional time In Europe in 2014.

Figure 5‐7 suggests a high level of capacity‐related airport arrival ATFM delays while the analysis in Figure 5‐10 confirms the application of capacity‐related ATFM regulations at the same time almost every day (right side) during the two daily arrival peaks (left side).

Whereas a number of regulations in the evening were linked to the use of a special runway configuration due to environmental reasons, the majority of regulations during the two arrival peaks were related to aerodrome capacity.

Possibilities of how the situation could be improved should be investigated as the systematic daily use of capacity related ATFM arrival regulations during the same time should not take place at coordinated airports. Investigations are on‐going concerning the analysis of the impact of incurred arrival regulations on the scheduled operations. In particular, to what extent systemic shares of the delay are incorporated into the schedule.

Figure 5‐10: Capacity‐related ATFM regulations at Zurich airport

After virtually no delay in 2013, Istanbul Sabiha Gökçen airport accounted for 7.4% of arrival ATFM delays in Europe in 2014. Figure 5‐11 shows the evolution of the average daily arrivals together with the incurred arrival ATFM delay between 2008 and 2014.

Figure 5‐11: Traffic and arrival ATFM delays at Istanbul Sabiha Gökçen airport (2008‐2014)

Traffic at Istanbul Sabiha Gökçen airport grew by +24% in 2014 (see also Figure 5‐1) with an impressive average annual growth rate of +27% between 2008 and 2014. In view of the ongoing growth and notable increase in airport arrival ATFM delays, the situation at the airports needs to be monitored.

While London Heathrow, Zurich and Istanbul Sabiha Gökçen airports were only three examples with a notable network impact in terms of arrival flow management performance, data for all the other airports is available in Annex VII.

0

5

10

15

20

25

30

35

40

45

50

0

5

10

15

20

25

30

35

40

45

50

0600

0645

0730

0815

0900

0945

1030

1115

1200

1245

1330

1415

1500

1545

1630

1715

1800

1845

1930

2015

2100

2145

2230

Local time

Hourly arrival rate at Zurich (ZRH) airport in 2014


Declared peakarrival capacity

360

420

480

540

600

660

720

780

840

900

960

1020

1080

1140

1200

1260

1320

01/01/2014

01/02/2014

01/03/2014

01/04/2014

01/05/2014

01/06/2014

01/07/2014

01/08/2014

01/09/2014

01/10/2014

01/11/2014

01/12/2014

01/01/2015

Time of day

06:00

08:00

10:00

12:00

14:00

16:00

18:00

20:00

22:00

Daily capacity‐related (codes G,V,C,S) airport arrival ATFM regulations at Zurich airport in 2014


0

50

100

150

200

250

300

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

JAN

MAR

MAY

JUL

SEP

NOV

JAN

MAR

MAY

JUL

SEP

NOV

JAN

MAR

MAY

JUL

SEP

NOV

JAN

MAR

MAY

JUL

SEP

NOV

JAN

MAR

MAY

JUL

SEP

NOV

JAN

MAR

MAY

JUL

SEP

NOV

JAN

MAR

MAY

JUL

SEP

NOV

2008 2009 2010 2011 2012 2013 2014

average arrivals per day

Avg. ATFM arrival delay per arrival

Traffic and ATFM arrival delay at Istanbul Sabiha Gokçen airport (2008‐2014)

Other (all other codes)

Weather (C,D)

ATC other (codes IRVT)

Cap./ staffing (codes CSG)

Avg. daily arrivals


62

Taxi‐in efficiency

Although a possible approach to measuring taxi‐in efficiency at airports was briefly outlined in PRR 2013 [Ref. 2], little focus has been put on the taxi‐in phase as it is considered to be of a lower order of magnitude from a performance point of view.

Moreover, the taxi‐in phase is affected by a number of factors most of which cannot be directly influenced by ATM. The main influencing factors are considered to be airport layout (runways, taxiways, stands, crossings, etc.) and stand allocation/availability.

For completeness reasons and to provide a better understanding of the order of magnitude of the level of inefficiencies estimated in the taxi‐in phase, this section provides an analysis based on a statistical method.

Figure 5‐12 shows the average additional taxi‐in times observed at the top 30 European airports in 2014.

Additional taxi‐in time

The Additional Taxi‐in time aims at evaluating the level of inefficiencies in the taxi‐in phase.

The analysis refers to the period between the time when the aircraft landed and the time it arrives at the stand. For each arrival, the additional time is computed as the difference between its actual taxi‐in time and the reference taxi‐in time based on the 20th percentile of the associated stand‐runway combination. In case the actual taxi‐in time is equal or less than the reference taxi‐in time, the additional time is set to zero.

Figure 5‐12 shows the average additional taxi‐in times observed at the top 30 European airports in 2014. On average, the additional taxi‐in time in 2014 is 1.6 minutes per arrival which is considerably lower than the taxi‐out inefficiencies observed in previous years (see also next section).

Figure 5‐12: Average additional taxi‐in time (2014)

As can be expected, the airports with the highest traffic volume, such as Frankfurt, London Heathrow and Paris Charles de Gaulle have a higher than average additional taxi‐in time.

Rome Fiumicino and Manchester have an additional taxi‐in time higher than Frankfurt which is most likely linked to stand limitations, especially during the summer period.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

London* (LHR)

Paris* (CDG)

Frankfurt* (FRA)

Amsterdam

(AMS)

Istanbul A

taturk (IST)

Munich* (M

UC)

Madrid* (M

AD)

Rome* (FCO)

Barcelona (BCN)

London* (LGW)

Zurich* (ZRH)

Copenhagen (CPH)

Oslo* (OSL)

Vienna (VIE)

Paris Orly (ORY)

Stockholm

(ARN)

Brussels* (BRU)

Duesseldorf* (DUS)

Gen

eva (GVA)

Berlin

Tegel (TXL)

Istanbul (SA

W)

Dublin

(DUB)

Antalya (AYT)

Palma (PMI)

Manchester (MAN)

Helsinki* (HEL)

Milan* (MXP)

Lisbon (LIS)

London (STN)

Athen

s (ATH

)

Avg. additional taxi in tim

e

2014

2013

Avg. (2014)

Source: PRC analysis* A‐CDM implemented airport


63

Factors considered for the take‐off sequence

TOBT

Stand, VTT, SID, AC type, wake vortex

ATFM slot

Departure capacity/ minimum departure interval, remote holds

Max. waiting time at runway

RWY config. / RWY in use

Optimised TOT

TSAT

‐ variable TXOUT

5.6 Management of the departure flow

Airport Collaborative Decision Making (A‐CDM) is considered to be a key enabler to increase the efficiency and predictability of the turn‐round phase and to better manage the taxi‐out phase in order to optimise the

departure sequence at the runway.

The real time sharing of milestones such as the target off block time (TOBT)33 and the resulting higher level of accuracy enables a better optimisation of the departure sequence in order to keep queuing at the runway and subsequent fuel burn to a minimum while maximising runway throughput. A TOBT prediction can be made by either the aircraft operator (AO) or delegated to ground handler (GH). The TOBT represents a commitment in terms of timeliness of operation. Ground handlers play a vital role in ensuring that the processes are completed and an aircraft can immediately push‐back or start‐up upon reception of the appropriate ATC clearance.

The Target Start‐up Approval Time (TSAT)34 computation is usually a component of the Departure Manager (DMAN) and is supplied and managed by the ANSP. The TSAT should reduce queuing times at the runway hold while maintaining a high runway utilisation. Based on the TOBT, the TSAT computation may consider ATFM departure slots, wake vortex, Standard Instrument Departure (SID) routing, variable taxi times based on stand runway combination, low visibility procedures and other factors.

In case of an AFTM regulated flight, local ATC will allocate a TSAT to meet the calculated take‐off time (CTOT).

ATFM departure slot adherence

The adherence to the ATFM departure slot is important to increase traffic flow predictability and to ensure that traffic does not exceed regulated capacity en‐route or at the destination airport. An ATFM slot tolerance window [‐5 min, +10 min] is available to ATC to sequence departures.

Figure 5‐13 shows the share of ATFM regulated departures at the top 30 airports in Europe (grey bar) and also the percentage of ATFM regulated departures outside the ATFM tolerance window (red line).

Overall, the number of ATFM regulated departures and the slot adherence decreased notably between 2010 and 2013 at the top 30 airports in Europe.

ATFM slot adherence

ATFM departure slots are allocated centrally by the Network Manager to hold aircraft on the ground when there is an envisaged imbalance between demand and capacity at airports or en‐route.

ATFM slot adherence measures the share of take‐offs outside the allocated ATFM window. Within the EU, the monitoring of ATFM slot adherence is required by Regulation 255/2010 [Ref. 23]. ATS units are required to provide information on non‐compliance for airports where non‐adherence equals or exceeds 20% of regulated departures.

Despite the increase of ATFM regulated departures from 8.2% in 2013 to 9.1% in 2014 at the top 30 airports in Europe, the share of flights outside the 15 minute tolerance window decreased further from 11.3% in 2013 to 10.4% in 2014.

33 The time that an aircraft operator or ground handler estimates that an aircraft will be ready, all doors closed,

boarding bridge removed, push back vehicle available and ready to start up/push back immediately upon reception of clearance from the TWR.

34 The time provided by ATC taking into account TOBT, CTOT and/or the traffic situation that an aircraft can expect start up/push back approval.

Turn‐round phase

(efficiency & predictability)

Departure flow management

ATC‐related departure delay

Taxi‐out additional

time

Optimisation of departure sequencing


64

Figure 5‐13: ATFM slot adherence at airports (2014)

The three Turkish airports showed the highest share of ATFM regulated departures outside the 15 minute tolerance window. It is interesting to note that Paris Charles de Gaulle and London Heathrow35 have a comparatively high share of departures outside the tolerance window, despite full A‐CDM implementation and departure sequencing. Airports should strive to continue the positive trend observed over the past years in order to further improve traffic flow predictability within the European network.

ANS‐related inefficiencies on the departure flow

Although the ability of ANS in reducing total time inefficiencies is limited when runway capacities are constraining departures, the goal should be to minimise inefficiencies of the departure process (e.g. apron, taxiway, and threshold queuing) by keeping aircraft longer at the stand.

In this respect, the optimisation of the departure queue management through the allocation of an optimised TSAT aims to maximise the runway throughput while keeping the additional fuel burn to the necessary minimum.

The efficiency of this balancing act can be measured by the additional taxi‐out time. Local ATC pre‐departure delay addresses the effect of capacity/demand imbalances surrounding the departure process.

Whereas the taxi‐out additional time is based on a statistical measure, the local ATC departure delay is derived from universally applicable IATA delay codes reported by airlines.

In particular, the analysis is based on the IATA delay code 89 which, besides delays caused by local ATC constraints, also includes delays due to late push‐back approval and some other reasons which may introduce a certain level of bias. Moreover, the delay attribution and coding is to some extent dependent on local and/or operator procedures and practices, and may vary across European airports.

Local ATC pre‐departure delay

Departure delays due to local ATC are a proxy for ATC induced delays at the departure stand as a result of demand/capacity imbalances in the manoeuvring area and/or terminal airspace.

Work is in progress to use additional sub‐codes to allow for a proper identification of the different causal factors. Further work revolves around the establishment of a local validation process at several airports to assert the usage of delay codes (including sub‐codes) identifying the right causal

35 An initiative to improve ATFM slot adherence at LHR launched at the end of 2014 should improve future

performance.

0%

5%

10%

15%

20%

25%

30%

35%

40%

London* (LHR)

Paris* (CDG)

Frankfurt* (FRA)

Amsterdam

(AMS)

Istanbul A

taturk (IST)

Munich* (MUC)

Madrid* (M

AD)

Rome* (FCO)

Barcelona (BCN)

London* (LG

W)

Zurich* (ZRH)

Copenhagen (CPH)

Oslo* (OSL)

Vienna (VIE)

Paris Orly (ORY)

Stockholm

(ARN)

Brussels* (BRU)

Duesseldorf* (DUS)

Gen

eva (GVA)

Berlin

Tegel (TXL)

Istanbul (SA

W)

Dublin

(DUB)

Antalya (AYT)

Palma (PMI)

Manchester (M

AN)

Helsinki* (HEL)

Milan* (M

XP)

Lisbon (LIS)

London (STN)

Athen

s (ATH

)

% of ATFM regulated departures

% of ATFM regulated departures outside the ATFM tolerance window

Average (ATFM regulated departures outside the ATFM tolerance window)

2014


0%

5%

10%

15%

20%

25%

30%

35%

40%

2010

2011

2012

2013

2014

Source: NM; PRU analysis

Top 30 average


65

factor. However, it is unlikely that these activities will be implemented consistently at all European airports within the near future.

A possible solution to overcome this shortcoming at A‐CDM airports could be the use of data from the A‐CDM milestone approach which captures the difference of the target and actual off‐block and start‐up times. In the absence of an ATFM regulation, a target start‐up approval time (TSAT) different from the TOBT signals a local ATC decision to hold the aircraft at the stand. Further analysis of the correlation of between the TSAT‐TOBT off‐set and reported delays, in particular code 89, is required.

Figure 5‐14 shows the local ATC departure delays (top of figure) and the taxi‐out additional time at the top 30 airports. Although the level of inefficiencies cannot be reduced to zero, on average, the less fuel efficient taxi‐out additional time is four times higher than the local ATC departure delays at the gate (3.5 vs. 0.9 minutes per departure) which suggests scope for further improvement.

Figure 5‐14: ANS‐related inefficiencies on the departure flow at the top 30 airports in 2014

Whereas the level of Local ATC departure delay remained almost constant, a slight reduction of additional taxi‐out time can be observed between 2010 and 2014.

Figure 5‐15: Airport contribution towards European network performance (Departure flow)

In additional to the average delay per departure shown in Figure 5‐14, Figure 5‐15 shows again the

0

2

4

6

8

10

London* (LHR)

Paris* (CDG)

Frankfurt* (FRA)

Amsterdam

(AMS)

Istanbul A

taturk (IST)

Munich* (M

UC)

Madrid* (M

AD)

Rome*

(FCO)

Barcelona (BCN)

London* (LGW)

Zurich* (ZRH)

Copen

hagen (CPH)

Oslo* (OSL)

Vienna (VIE)

Paris Orly (ORY)

Stockholm

(ARN)

Brussels* (BRU)

Duesseldorf* (DUS)

Gen

eva (GVA)

Berlin Tegel (TXL)

Istanbul (SA

W)

Dublin

(DUB)

Antalya (AYT)

Palma (PMI)

Manchester (M

AN)

Helsinki* (HEL)

Milan* (M

XP)

Lisbon (LIS)

London (STN)

Athen

s (ATH

)

min. per dep

arture

2014

2013

0

2

4

6

8

10

2014

2013

Local ATC pre‐departure delays (at gate)

0

1

2

3

4

5

0

1

2

3

4

5

2010

2011

2012

2013

2014

Taxi‐out additional time

Source: NM; PRC

not available

not available

not available

not available


0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

London* (LHR)

Paris* (CDG)

Frankfurt* (FRA)

Amsterdam

(AMS)

Istanbul A

taturk (IST)

Munich* (MUC)

Madrid* (MAD)

Rome*

(FCO)

Barcelona (BCN)

London* (LGW)

Zurich* (ZRH)

Copen

hagen (CPH)

Oslo* (OSL)

Vienna (VIE)

Paris Orly (ORY)

Stockholm

(ARN)

Brussels* (BRU)

Duesseldorf* (DUS)

Gen

eva (GVA)

Berlin

Tegel (TXL)

Istanbul (SA

W)

Dublin

(DUB)

Antalya (AYT)

Palma (PMI)

Manchester (M

AN)

Helsinki* (HEL)

Milan* (M

XP)

Lisbon (LIS)

London (STN)

Athens (ATH

)

% of total (top 30 airports)

Airport contribution towards total network performance (departure flow)

Taxi out

Local ATC departure delays



66

airports’ contribution towards European network performance in terms of the share of total delays (top 30 airports).

Figure 5‐16 shows the variation of the additional taxi‐out times for Rome Fiumicino. Stark seasonal variations of the additional taxi‐out times can be observed. Next to the additional taxi‐out time, Rome shows a reasonable high level of ATC pre‐departure delay of above 2 minutes per flight (see Figure 5‐14).

Figure 5‐16: Additional Taxi‐out times at Rome Fiumicino airport

Rome Fiumicino became a fully implemented A‐CDM airport in March 2014. In order to optimise the departure sequencing at Rome Fiumicino airport, the information available through the A‐CDM processes should enable to identify the constraining factors (e.g. apron, gates, taxiways, or runways).

De‐icing operations

An important factor affecting departure sequencing is the need of de‐icing operations which may become the bottleneck for the departure flow. This causes an issue when preparing the take‐off sequence as the de‐icing operations and the required de‐icing time may vary dependent on the airport, stand (on‐stand/remote de‐icing), and the actual weather and traffic conditions. This additional element must be built into the calculation of the TOBT and TTOT and be considered when establishing the appropriate departure sequence.

For stand de‐icing, the TOBT needs to incorporate the de‐icing time to provide an accurate off block time. For remote de‐icing, there is no change to the TOBT and TSAT processes. However, the TTOT needs to be revised to include the required additional time for taxiing to the remote de‐icing position, the de‐icing time, and subsequent taxiing to the departure runway.

Figure 5‐17 shows the 30 day moving average for taxi‐out additional time at Munich (left) and Vienna airport (right).

Figure 5‐17: Impact of winter operations on additional taxi times

Winter 2013/14 was clearly better than the previous one and winter operations had a notable impact

0

1

2

3

4

5

6

7

8

9

1001/10/2011

01/12/2011

01/02/2012

01/04/2012

01/06/2012

01/08/2012

01/10/2012

01/12/2012

01/02/2013

01/04/2013

01/06/2013

01/08/2013

01/10/2013

01/12/2013

01/02/2014

01/04/2014

01/06/2014

01/08/2014

01/10/2014

01/12/2014Average additional taxi out time (m

ins)

Avg. daily additional time

Rolling average (30 day)

Avg. daily additional taxi out time at Rome (FCO) airport


0

1

2

3

4

5

6

7

8

9

10

01/10/2011

01/12/2011

01/02/2012

01/04/2012

01/06/2012

01/08/2012

01/10/2012

01/12/2012

01/02/2013

01/04/2013

01/06/2013

01/08/2013

01/10/2013

01/12/2013

01/02/2014

01/04/2014

01/06/2014

01/08/2014

01/10/2014

01/12/2014

Average additional taxi out time (mins)



Avg. daily additional taxi out time at Munich (MUC) airport


0

1

2

3

4

5

6

7

8

9

10

01/10/2011

01/12/2011

01/02/2012

01/04/2012

01/06/2012

01/08/2012

01/10/2012

01/12/2012

01/02/2013

01/04/2013

01/06/2013

01/08/2013

01/10/2013

01/12/2013

01/02/2014

01/04/2014

01/06/2014

01/08/2014

01/10/2014

01/12/2014

Average additional taxi out time (mins)



Avg. daily additional taxi out time at Vienna (VIE) airport



67

on taxi‐out time and on the year‐on‐year change of the average additional taxi‐out time. This highlights the need to better consider de‐icing operations when evaluating taxi‐out performance at airports.

5.7 Conclusions

In 2014, IFR movements at the top 30 airports increased by +2.7% compared to the previous year, which is stronger than the European traffic growth in 2014 (+1.7% vs. 2013). Despite the growth in 2014, traffic at the top 30 airports remains 3.3% below 2008 levels. Compared to 2013, the number of passengers at the top 30 airports increased by +5.8% in 2014 (9.9% above 2008 levels).

There were notable differences in growth between airports in 2014. The Turkish airports continued their remarkable growth already observed in previous years and, following a traffic decrease at Paris Charles de Gaulle and Frankfurt airport, London Heathrow now ranks first in terms of average daily movements in Europe.

Operational cancellations at airports were included for the first time in this year’s report. Overall, around 1.5% of the flights confirmed by the air carrier the day before operations were cancelled in 2014. Data on cancellations were not available from all airports and thus limit the scope of the analysis of this report. With the on‐going establishment of the airport operator data flow specification, the level of reporting will improve in the future.

In order to avoid excessive disruptions in daily operations, at larger airports is already regulated strategically in terms of volume and concentration through the airport capacity declaration process. On the day of operations, ANS plays a key role in balancing traffic with available capacity at airports within the given infrastructural and environmental constraints and in the integration of airports in the European network. In view of the considerable downward adjustment of airport expansion plans in Europe, the optimised use of available capacity is crucial to keep delays to a minimum.

The indicators used for the evaluation of ANS‐related performance relate to the management of the inbound (i.e. arrival ATFM delay and ASMA additional time) and outbound (i.e. ATC pre‐departure delay and additional taxi out time) traffic flow and are consistent with the SES performance scheme.

In 2014, the average additional time for ASMA and taxi‐out improved slightly at the top 30 European airports whereas airport ATFM arrival delays and ATC pre‐departure delays increased at the same time.

London Heathrow, Frankfurt, Amsterdam Schiphol, Rome Fiumicino, Zurich, and the two Istanbul airports had the most notable impact on the network in 2014. The underlying reasons differ by airport and range from the airport scheduling process to the ability to sustain declared capacity in various circumstances on the day of operations.

Improved data exchange fostered by A‐CDM implementation and increased situation awareness are key enabler for performance improvements at airports. However the mere availability of information does not automatically resolve issues without proactively addressing relevant optimisation processes.

For example, initial work on XMAN demonstrated positive improvements in term of reducing the additional ASMA time for London Heathrow. The better use of cross‐border arrival management (linking the terminal approach and en‐route function, is a key enabler for the absorption of possible holdings in a more environmental friendly manner, as the absorption takes place in the en‐route phase through reduced cruise speed.

On the other hand, local constraints at Rome Fiumicino pose a challenge to take advantage of the benefits of A‐CDM with a view to the optimisation of departure sequencing process.

The evaluation of daily taxi‐out performance shows that the current performance framework does not sufficiently address the impact of adverse weather on the performance indicators. Future work is required in order to better evaluate the level of efficiency in such conditions (e.g. de‐icing conditions).

Chapter 6: ANS Cost‐efficiency

PRR 2014 ‐ Chapter 6: ANS Cost‐efficiency 68

6 ANS Cost‐efficiency

KEY POINTS KEY DATA 2013 vs. 12

EN‐ROUTE ANS (EUROCONTROL AREA) En‐route ANS costs

Real en‐route unit cost decreased in 2013 (a reduction of ‐3.3% compared to 2012).

At system level, 2013 was a year of moderate SUs growth (+2.1%), following a decrease in 2012.

En‐route ANS costs (expressed in €2009) decreased overall by ‐1.3%.

Total en‐route ANS costs (M€2009) 6 426 ‐1.3%

Service units (M) 120.8 +2.1%

En‐route ANS costs per SU (€2009) 53.2 ‐3.3%

TERMINAL ANS (SES RP1 AREA) Terminal ANS costs

Terminal ANS costs information differs across States and across time for many reasons, although quality and quantity of data is gradually improving.

For the purpose of this analysis, terminal navigation service unit (TNSU) were recomputed using the formula mandatory from 2015.

For SES States (RP1), in 2013, both terminal ANS costs (‐3.9%) and unit costs (‐3.6%) decreased in real terms for the fourth year in a row.

Total terminal ANS costs (M€2009) 1 354 ‐3.9%

Recomputed terminal service units ((MTOW/50)^0.7) (M)

7.7 ‐0.3%

Terminal ANS costs per terminal SU (€2009) 175.4 ‐3.6%

GATE‐TO‐GATE ANSP Gate‐to‐gate ATM/CNS provision costs

In 2013, ATM/CNS provision costs reduced by ‐2.0% while the number of composite flight‐hours remained fairly constant. As a result, unit ATM/CNS provision costs decreased by ‐1.9% in 2013 to reach an amount of €436.

In parallel, the unit costs of ATFM delays significantly fell (‐18.2%) for the third year in a row. In 2013, unit economic costs (which also consider the cost of delay) amounted to €478, which is ‐3.6% lower than in 2012 and ‐13.0% lower than pre‐recession levels (€549 in 2008).

In fact the unit economic costs reached in 2013 their lowest level since monitoring this metric in 2001.

Gate‐to‐gate ATM/CNS provision costs (M€ 2013)

7 937 ‐2.0%

Composite flight‐hours (M)

(see page 86 for definition of composite flight hours)

18.2 ‐0.1%

Gate‐to‐gate ATM/CNS provision costs per composite flight‐hour (€ 2013)

436 ‐1.9%

6.1 Introduction

This chapter analyses ANS cost‐efficiency performance in 2013 (i.e. the latest year for which actual financial data are available) and provides a performance outlook, where possible.

This chapter provides a pan‐European view, covering both the 29 States which are subject to the requirements of the SES Performance Scheme (“SES States”) and the 9 non‐SES States which are members of EUROCONTROL (see section 6.2 below).

The cost‐efficiency performance of SES States in 2013 has already been scrutinised in accordance with the SES Regulations and the results published in the PRB's monitoring report in October 2014 [Ref. 24]. This present report does not seek to duplicate this analysis nor assess in any way performance against SES targets. Instead, it takes the SES data and aggregates it with the data for the non‐SES States to reach a pan‐European view. However some SES States have updated their data since the PRB report was published. In order to reflect the best available information on States cost‐efficiency performance, this report uses the most up‐to‐date data.

This chapter also considers the outlook for 2015‐2019. Again, for SES States the data is taken from what States have already provided, namely the latest available data provided by SES States up to 15 January 2015. As with data for 2013, the SES data is aggregated with that for non SES States to reach a pan‐European view.

For transparency purposes, the reconciliation with the figures reported in the PRB 2013 monitoring report and PRB RP2 assessment report [Ref. 25] is given in Annex VIII (page 101 of this PRR).

PRR 2014 ‐ Chapter 6: ANS Cost‐efficiency

69

Determined cost method

In the “determined costs method”, the amounts charged to airspace users for a given year (N) are now fixed prior to the RP (the “determined costs”). The difference between actual costs and determined costs is borne/retained by the State/ANSP concerned and the difference in revenues due to the difference between actual traffic and traffic forecasted prior to the period for that year is shared between ANSPs and airspace users. This method is expected to drive the ANSPs behaviour to reduce their costs in the case of lower traffic than planned and the other entities (State, NSAs, MET service providers) to contain the actual costs within the determined costs envelope. This means that the actual unit costs slightly differ from the amounts ultimately paid by the airspace users.

Section 6.2 presents a detailed analysis of en‐route cost‐efficiency performance in the EUROCONTROL area, including a new sub‐section on the en‐route unit cost ultimately incurred by airspace users for 2013 (sometimes also referred to as the “true cost for users”). Section 0 gives an evaluation of terminal ANS costs within the Single European Sky (SES) area.

In order to ensure consistency and comparability with indicators defined in the Single European Sky (SES) performance scheme and the information provided in national/FAB Performance Plans, the cost‐efficiency indicators in Sections 6.2 and 0 are expressed in terms of costs per service unit (SU) and in Euro 2009.

Finally, Section 6.4 provides a factual benchmarking analysis of ANSPs’ gate‐to‐gate economic performance focusing on ATM/CNS costs under direct ANSP responsibility and estimated cost of delay attributable to the respective service providers.

6.2 En‐route ANS cost‐efficiency performance

The analysis of en‐route ANS cost‐efficiency in this section refers to the en‐route charging zones in EUROCONTROL's Route Charges System in 2013 (with the exception of the Portugal Santa Maria charging zone) plus Estonia (which joined EUROCONTROL on 1 January 2015).

The “RP1 SES States” refer to the 27 Member States of the European Union included in the 1st reference period (RP1) of the SES performance scheme, plus Switzerland and Norway.

Croatia, which joined the EU in July 2013, is not included in the “RP1 SES States” in the analyses in this chapter. For the RP1 SES States, operating in the context of the SES Regulations, 2013 is the second year in which the “determined costs” method with specific risk‐sharing arrangements, defined in the Charging Regulation [Ref. 26] aiming at incentivising economic performance, is applied.

“RP1 non‐SES States“ refers to the other nine EUROCONTROL States participating in the Route Charges System in 2013 (i.e. Albania, Armenia, Bosnia–Herzegovina, Croatia, FYROM, Moldova, Serbia, Montenegro and Turkey). For these nine States, the “full cost‐recovery method” continued to apply in 2013. The “RP1 SES States” and “RP1 non‐SES States“ are shown in Figure 6‐1).

In order to evaluate possible differences in trends and behaviour between those states operating in the context of the SES Regulations and the other states in the Route Charges System, the results of the 2013 analysis are presented separately for the SES States and non‐SES States.

Figure 6‐1: SES States (RP1) and non‐SES States in RP1

SES states

Non SES states

SES and non SES states


70

Trends in actual en‐route cost‐efficiency performance

Figure 6‐2 provides a summary view of the actual en‐route cost‐effectiveness data between 2009 and 2013 and the forecast for 2014, including the changes in the en‐route ANS costs per SU.

Figure 6‐2: Real en‐route unit costs per SU for EUROCONTROL Area (€2009)

In 2013, at pan‐European level actual en‐route ANS costs decreased by ‐1.3% while traffic (en‐route SUs) increased by +2.1%. As a result, actual en‐route unit costs in 2013 decreased by ‐3.3% compared to 2012. At this stage it is important to note that, although the 2013 traffic is +2.1% higher than in 2012, it is significantly lower than the expectations during the performance planning period back in 2011.

Trends in actual costs by nature (2013 vs. 2012)

Figure 6‐3 shows the trends in total en‐route costs broken down by nature.

The largest cost decrease in 2013 is observed for staff costs (‐81.1M€2009) driven by RP1 SES States.

Figure 6‐3: Difference between 2013 and 2012 costs by nature (€2009)

The majority of States (17 SES charging zones and 3 non‐SES charging zones) show a decrease in staff costs compared to 2012. Decreases in four specific States have a significant impact at system level:

Sweden: ‐36.6M€2009 (‐29.1%). This decrease in 2013 is mainly due to fact that Sweden 2012 en‐route cost‐base included significant exceptional costs relating to pensions and also reflects a reduction in the number of staff.

60.1 56.7 53.8 55.0 53.2 53.2

97 97 98 97 100

103

108 107 109

113

20

30

40

50

60

2009Actuals

2010Actuals

2011Actuals

2012Actuals

2013Actuals

2014Forecasts

En‐route real cost

per SU (€2009)Real en‐route unit

costs per SU for the EUROCONTROL

Area (€2009)




2009

Actuals

2010

Actuals

2011

Actuals

2012

Actuals

2013

Actuals

2014

Forecasts

2013 vs

2012

2014 vs

2013

2009‐14

AAGR

Total en‐route ANS costs (M€2009) 6 648 6 477 6 453 6 510 6 426 6 640 ‐1.3% 3.3% 0.0%

SES States (EU‐27+2) 6 248 6 069 5 972 6 048 5 948 6 155 ‐1.6% 3.5% ‐0.3%

Other 9 States in the Route Charges System 400 407 481 463 478 485 3.4% 1.4% 3.9%

Total en‐route service units (M SU) 111 114 120 118 121 125 2.1% 3.4% 2.5%

SES States (EU‐27+2) 98 100 105 104 105 109 1.6% 3.4% 2.1%

Other 9 States in the Route Charges System 13 14 15 15 16 16 5.2% 3.2% 5.2%

En‐route real unit cost per SU (€2009) 60.1 56.7 53.8 55.0 53.2 53.2 ‐3.3% 0.0% ‐2.4%

SES States (EU‐27+2) 63.7 60.4 56.9 58.4 56.6 56.6 ‐3.2% 0.1% ‐2.3%

Other 9 States in the Route Charges System 31.9 29.6 32.5 31.1 30.6 30.0 ‐1.8% ‐1.8% ‐1.2%

‐2.1%

‐0.9%

1.2%

‐6.3% ‐1.3%

‐2.5%

‐1.4%

0.9%

‐6.5% ‐1.6%

4.3%

3.5%

6.0%

‐4.7%

3.4%

‐20.0%

‐15.0%

‐10.0%

‐5.0%

0.0%

5.0%

(120)

(100)

(80)

(60)

(40)

(20)

‐

20

40

Staff Otheroperatingcosts

Depreciation Cost ofcapital

Exceptionalitems

Total costs

€2009 (million)

Total

SES States

non‐SES States



71

Staff costs and provisions for future obligations

Large variations in staff costs may occur as a result of variations in (accounting) provisions for future obligations. The volatility of the (accounting) provisions raises concerns in the context of charging and performance, whereas increases or decreases in these provisions do not necessarily represent costs directly attributable to the provision of ANS in the year in which they are recorded. Moreover, these increases or decreases in provisions ‐especially when related to (Defined Benefits) pensions scheme ‐ can be significant in size and thereby influence the resulting cost‐efficiency indicator, which may no longer reflect the adjustment of costs to the traffic context and the genuine cost‐efficiency performance of States/ANSPs or even the Pan‐European system as a whole. For those States under the determined costs method, these increases may also significantly impact the future amounts charged to airspace users if deemed eligible as exemptions from cost‐sharing in accordance with the SES Charging Regulation.

Spain (Continental and Canarias): ‐22.5M€2009 (‐5.5%), mainly reflecting the impact of the Social Plan for Voluntary Lay‐offs adopted in 2012 which led to the departure of 249 non‐ATCO staff in the first half of 2013.

Germany: ‐17.3M€2009 (‐2.8%) due to a decrease in the number of staff in DFS payroll. It is also noteworthy that the decrease in staff costs in 2013 follows an increase of +57.3M€2009 in 2012 (+10% compared to 2011) resulting from an increase in pension costs consequent to a change in the discount rate for occupational pensions and higher gross wages and salaries.

UK: ‐13.6M€2009 (‐4.9%) driven by “pay restraints and lower headcount”. The actual staff costs reported by NERL for the year 2013 do not include the accounting pension contributions as reported under IFRS but comprise regulatory pension allowances, which are higher. Taking into account the accounting pension costs instead of the regulatory allowances would lead to lower actual staff costs for NERL in 2013. It should also be noted that the UK reported +55.6M€2009 of exceptional costs in 2013, i.e. an increase of +38.6M€2009 “resulting from NERL’s voluntary redundancy programme in 2013 for over 240 people from all parts of NERL’s business.” Considering these additional exceptional costs as staff costs, the increase in UK staff costs represent an increase of +25.06M€2009 (+14.4% compared to 2012).

Other significant decreases in staff costs in percentages (above ‐10%) are observed in Serbia and Montenegro (‐4.6M€2009 or‐13.3%); Portugal (‐11.1 M€2009 or‐13.2% compared to 2012) and Finland (‐2.9M€2009 or‐12.7%).

The most notable increases in staff costs in absolute terms in 2013 are observed in France (+12.7 M€2009 or +2.0%); Norway (+10.9 M€2009 or +17.6%) and Turkey (+8.6 M€2009 or +7.1%).

Other significant increases in staff costs in relative terms (above +10% compared to 2012) are observed in Bosnia and Herzegovina (+27.3%); Moldova (+23.0%); Albania (+11.9%) and Malta (+10.7%).

Other operating costs in 2013 show an overall decrease of ‐14.2 M€2009 or ‐0.9% compared to 2012.

The main decreases in other operating costs in 2013 are observed in Spain (‐12.5M€2009 or ‐10.3% compared to 2012 for the two charging zones), mainly reflecting cost‐saving measures implemented by Aena, including renegotiation of contracts, reduction in rentals and reduction of external consultants, as well as measures of efficiency to reduce consumption of supplies. To note that this follows a decrease of ‐20.9M€2009 in 2012 compared to 2011 (or ‐14%).

Other significant decreases in percentages (above ‐10% compared to 2012) are observed in Serbia and Montenegro (‐23.4%); Albania (‐20.7%); Slovakia (‐17.6%); Armenia (‐16.6%); Moldova (‐13.8%); Denmark (‐13.4%); Latvia (‐12.2%); Hungary (‐11.9%) and Switzerland (‐10.7%).

The decreases in 2013 in other operating costs were in part offset by the increases in Belgium‐Luxembourg (+8.6M€2009 or +37.7%, mainly relating to a “provision for legal matters (lawsuit)”); Sweden (+8.2M€2009 or +13.7%) and Turkey (+7.8 M€2009 or +8.6%). Significant percentage increases (above +10%) in 2013 were observed in Finland (+2.3M€2009 or +17.8%); Croatia (+2.5M€2009 or +15.6%) and FYROM (+0.2M€2009 or +10.4%).

Depreciation costs increased by +1.2% overall in 2013, suggesting a potential catch‐up effect following the ‐1.6% decrease observed in 2012.


72

After a significant increase in the RP1 SES States in 2012 (+8%), as a result of an increase in the return on equity set in the context of the “determined costs method” and the new traffic risk‐sharing arrangements, the cost of capital decreased in 2013, mainly reflecting lower rates of interest on debt and/or lower actual asset bases, given lower than expected investments.

The main variation in exceptional items relates to the UK (NERL’s voluntary redundancy programme in 2013, see staff costs above).

Trends in actual unit costs by State/charging zone (2013 vs. 2012)

The bottom of Figure 6‐4 below shows the actual unit cost for each individual State (charging zone) in 2013. It ranges from 73.5€2009 in Germany to 19.8€2009 in Estonia, a factor of more than three. Figure 6‐4 also presents the changes in SUs and costs compared to 2012.

While it is to be expected that States’ unit costs vary due to, for example, different complexity and different economic factors36, some of the observed differences in Figure 6‐4 are more than expected after taking these factors into account. This raises questions about the cost‐effectiveness of some ‘high cost’ providers. In this respect, it should be highlighted that:

Some States appear in the “high unit cost” half of Figure 6‐4, but this result does not seem to tally with the prevailing complexity and economic environment (e.g. FYROM, Albania);

Some States, which were already flagged as being “high unit cost” have further increased their unit cost for providing the service (e.g. Slovenia);

Bosnia and Herzegovina has recorded a +15% increase in its unit cost which is the most significant increase across the EUROCONTROL Member States. Its en‐route ANS costs cumulate the costs for the national ANSP (BHANSA) which has started the provision of en‐route services in the airspace of the FIR only in 2014 (over parts of the airspace and up to 32 500 feet), as well as costs for the services provided by the ANSPs of Serbia and Croatia.

Figure 6‐4: 2013 Real en‐route ANS costs per SU by charging zone (€2009)

The efforts made by Spain over the past few years are showing results, as the en‐route unit cost for

36 Those metrics are measured annually as part of the ACE Benchmarking Report.

73.5

73.2

72.5

69.5

67.6

66.4

65.8

64.9

64.4

60.1

59.1

58.3

55.0

54.6

53.3

51.5

51.3

45.6

43.4

42.0

40.6

40.2

36.8

35.1

35.1

35.0

34.9

33.0

32.3

32.2

30.2

27.3

26.6

26.4

25.2

20.1

19.8

53.2

Germany

United

Kingdom

Switzerland

Italy

Spain Continental

Austria

Slovenia

Belgium/Lux.

Spain Canarias

France

Den

mark

Netherlands

FYROM

Sweden

Slovakia

Norw

ayFinland

Lithuania

Bosnia and Herzegovina

Albania

Croatia

Czech Republic

Hungary

Moldova

Portugal

Romania

Serbia and M

ontenegro

Greece

Cyprus

Bulgaria

Poland

Ireland

Turkey

Latvia

Arm

enia

Malta

Estonia

Pan

European

‐4%

6%

‐5% ‐2

%

‐7%

0%

7%

‐1%

0%

‐2%

‐6%

‐6%

‐1%

‐17%

‐5%

2%

‐1% ‐1%

15%

‐1%

6%

0%

‐3% 0%

‐11%

‐13%

‐11%

1%

‐4%

‐4%

‐5%

‐4% ‐2%

‐6% ‐2%

‐3% ‐2

%

‐3%

‐20%

‐15%

‐10%

‐5%

0%

5%

10%

15%

20%Actual unit cost (% change vs. 2012) Service units (% change vs. 2012) Actual en‐route costs (%change vs. 2012)

% chan

ge vs. 2012

Actual 2013 unit

cost (EU

R 2009)



73

Spain is now lower than the average of the other four largest States/ANSPs for the first time in 2013 (by ‐0.3% for Spain Continental and by ‐1.0% for Spain’s two charging zones together).

Trends in actual unit costs for RP1 SES States and “non‐SES” States (2013 vs. 2012)

Figure 6‐5 shows the evolution of key en‐route cost indices for RP1 SES States and non SES States.

Figure 6‐5: En‐route costs and service unit growth (RP1 SES States and non‐SES States)

The evaluation of differences in trends and behaviour between those States operating in the context of the SES Regulations and the other states in the Route Charges System does not yet show a clear cut trend and it is likely that a longer period would need to be considered. However, the following observations can be made at this stage.

In the RP1 SES States, overall:

After a decline in 2012 (‐1.5%), traffic in the SES area (in TSUs) increased again by +1.6% in 2013;

The 2013 average unit cost (56.6€2009) is significantly higher than for the non‐SES States (30.6€2009), reflecting different performance levels, and also the fact that they are operating in different economic and operational environments (e.g., generally higher cost of living and higher traffic complexity, in particular in the “Core Area”);

The trends and indicators are dominated by the “5 largest” States (France, Germany, Spain, Italy and the UK), accounting for 66.3% of the costs for the RP1 SES States in 2013 and for 55.4% in the number of TSUs);

The costs trend for the RP1 SES States is largely influenced by significant reductions in staff costs (‐81.1M€2009 see Figure 6‐3).

In non‐SES States,

Total costs for the non‐SES States increased by +3.4% in 2013. However this increase in costs occurred in the context of a higher traffic growth (+5.2%) resulting in a decrease of ‐1.8% in unit costs. The lower level of non‐SES unit costs compared to the RP1 SES States reflects lower complexity and lower cost of living, although the differences in both the cost of living and the complexity are reducing.

The trends and indicators are largely impacted by Turkey (accounting for 59.1% of the costs for the non‐SES States in 2013 and for 68.0% in the number of TSUs).

63.7 60.4 56.9 58.4 56.631.9 29.6 32.5 31.1 30.6

En route SUindex (2009)(RP1 SES States)

En route SUindex (2009)(Non‐SES)

En‐route costsindex (2009)

(RP1 SES States)

En‐route costsindex (2009)(Non‐SES)

80

85

90

95

100

105

110

115

120

125

130

20

25

30

35

40

45

50

55

60

65

2009(A) 2010 (A) 2011 (A) 2012 (A) 2013 (A)

En-

rout

e re

al c

ost

per

SU

(€2

009)


SES States

Non‐SES States


74

Actual 2013 versus 2013 plans/forecasts

Figure 6‐6 compares the forecast en‐route ANS costs and SUs prepared by the States for setting their 2013 en‐route unit rates with the actual costs and SUs provided by the States in November 2014. For the RP1 SES States, the forecasts en‐route ANS costs and SUs were determined as part of their adopted national/FAB Performance Plans for RP1 (i.e. 2011 and in some cases early 2012).

It is important to monitor the actual costs data against what was planned or forecasted for the year as it enables to evaluate the response to variations in traffic as well as the maturity of the planning process, two key elements for managing performance.

For the SES States (RP1), as highlighted in 6.2 above, the PRB has issued a detailed monitoring report for 2013, including an analysis of the ex‐ante and ex‐post profitability of the main ANSPs in respect of the activities performed in the year.

Figure 6‐6: Real en‐route ANS costs per SU, 2013 Actuals vs. Forecasts (in €2009)

Figure 6‐6 indicates that the actual real en‐route unit cost per service unit for 2013 is slightly lower than planned for 2013 (by ‐0.3%), as actual costs are ‐5.6% lower than expected in response to actual TSUs which were ‐5.3% lower than those considered for setting the 2013 unit rates.

Unit costs by State/charging zone (2013 actuals vs. 2013 plans/forecasts)

The difference in the planned and actual real en‐route unit costs for providing the service is provided at individual State level (charging zone) in Figure 6‐7.

Figure 6‐7: 2013 Real en‐route ANS costs per SU: Actuals vs. Forecasts (in €2009) by charging zone

€2009 prices Planned for 2013 Actuals 2013 Difference (%)

Total en‐route ANS costs (M€2009) 6 804 007 516 6 426 113 120 ‐5.6%

SES States (EU‐27+2) 6 318 609 442 5 947 883 649 ‐5.9%

Other 9 States in the Route Charges System 485 398 074 478 229 471 ‐1.5%

Total en‐route service units (M SU) 127 545 272 120 818 385 ‐5.3%

SES States (EU‐27+2) 111 461 030 105 171 670 ‐5.6%

Other 9 States in the Route Charges System 16 084 243 15 646 716 ‐2.7%

En‐route real unit cost per SU (€2009) 53.3 53.2 ‐0.3%

SES States (EU‐27+2) 56.7 56.6 ‐0.2%

Other 9 States in the Route Charges System 30.2 30.6 1.3%

Cost‐efficiency (2013)

‐97.2

‐59.0

‐42.5

‐38.1

‐22.9

‐15.6

‐14.6

‐13.3

‐13.2

‐10.7

‐9.4

‐8.4

‐8.2

‐7.7

‐6.3

‐6.1

‐4.5

‐4.0

‐3.4

‐2.7

‐2.1

‐1.8

‐1.7

‐1.0

‐0.6

‐0.6

‐0.2

‐0.2

0.2

0.2

0.4

0.6

0.7

0.8

0.9

5.1

9.4

Spain Continental

France

Italy

Germany

United

Kingdom

Austria

Poland

Greece

Ireland

Serbia and M

ontenegro

Den

mark

Bulgaria

Switzerland

Hungary

Spain Canarias

Belgium/Lux.

Czech Rep

ublic

Swed

en

Bosnia and Herzegovina

Finland

Portugal

Slovenia

Slovakia

Estonia

Latvia

Cyprus

FYROM

Arm

enia

Lithuania

Romania

Netherlands

Albania

Malta

Croatia

Moldova

Turkey

Norw

ay

‐3%

‐2%

1%

5% 6%

5%

‐10%

5%

‐9% ‐7% ‐7%

‐12%

2%

‐7%

8%

‐1%

‐3%

1%

‐2%

10%

2%

1%

‐4%

0%

‐4% ‐2%

5%

2%

1% 1% 1%

7%

‐16%

5%

1% 3%

‐4%

‐20%

‐15%

‐10%

‐5%

0%

5%

10%

15%

20%

25%

30%Difference in unit cost % Difference in costs in % Difference in SU in %

Difference in

costs

(Actual 2013 vs. Forecast

2013 in

M€2009)

Difference actuals 2013

vs. forecast 2013 (%)

(Actual 2013 vs. Forecast 2013)



75

As shown in the bottom of Figure 6‐7, the largest reductions in actual 2013 costs in value compared to what was forecasted are observed in:

Spain: mainly lower operating costs reflecting austerity measures implemented by Aena;

France: lower staff costs (containment); reduced depreciation as a result of lower actual capex than planned and change in applied accounting rules, and lower interest on debt;

Italy: mainly through lower staff costs (savings in salary and overtime, as well as new redundancy policy adopted in the end of 2012), and depreciation costs were mainly affected by postponed investments and renegotiated contracts with suppliers;

Germany: mainly lower staff costs reflecting a reduction in FTEs and therefore lower remuneration and social security expenses than planned. In addition, the actual pay increase in 2013 was +2% instead of +3% as foreseen;

UK: mainly in operating costs, reflecting costs reduction measures implemented by NERL, including reduction in support staff.

“True costs for users” for RP1 SES States and “non‐SES” States (2013 actuals vs. 2013 plans/forecasts)

In the determined costs method applied by the SES States, the costs ultimately charged to airspace users are no longer equivalent to the actual costs incurred by the States/ANSPs (as it is the case for the full cost recovery method applied by the non‐SES States).

From 2012 onwards, it is therefore important to monitor not only the performance of the States/ANSPs (actual costs incurred by the States/ANSPs for the activities performed in a given year) but also the amounts ultimately charged to the airspace users in respect of the activities of that year (sometimes also referred to as the “true cost for users”).

This objective of this section is to compare the amounts that were planned to be charged to the airspace users through the 2013 unit rates in respect of activities carried out in 2013 (determined costs for SES States and chargeable costs for non‐SES States), with the amounts that will ultimately be paid by the users in respect of these activities (“true cost for users”).

For the SES States:

In the “determined costs method”, the amounts charged to airspace users for a given year (N) are now fixed prior to the RP (the “determined costs”). The difference between actual costs and determined costs is borne/retained by the State/ANSP concerned and the difference in revenues due to the difference between actual traffic and traffic forecasted prior to the period for that year is shared between ANSPs and airspace users. This method is expected to drive the ANSPs behaviour to adjust their costs downwards when traffic is lower than planned and the other entities (State, NSAs, MET service providers) to contain the actual costs within the determined costs envelope. This is indeed what happened in 2013: actual SUs were ‐5.6% lower than forecast and in response the actual costs were ‐5.9% lower than planned (see Figure 6‐6 above).

Due to the traffic risk‐sharing, cost‐sharing and other adjustments provided in the Charging Regulation, the amounts ultimately paid by the airspace users differ from the actual costs. As a result, the “true costs for users” in respect of 2013 are some +205 M€2009 (+3.5%) higher than the actual costs of States/ANSPs. However these “true costs for users” (including all the adjustments) are some ‐166 M€2009 (‐2.6%) lower than the determined costs provided for 2013 in the RP1 performance plans.

In assessing the performance plans for RP2, the fact that several States/ANSPs demonstrated lower costs than planned was duly considered in order to ensure that users benefit from the cost reductions achieved during RP1.

For the non‐SES States:


76

In the “full cost recovery method”, the actual costs for the services provided in a given year (N) will ultimately be paid by the airspace users (through the unit rates for year N and adjustments to the unit rates of subsequent years). The “true costs for users” are therefore equal to the actual costs for the services provided in that year (after deduction of actual costs for exempted VFR flights and after deduction of actual other revenues).

Hence, the PRC computes that the “true costs for users” in 2013 are lower by ‐7.1 M€2009 compared to what was planned to be charged to users in respect of 2013 activities (see Figure 6‐6 above).

Pan‐European en‐route cost‐efficiency outlook for 2015‐2019

It should be noted that for RP2 the SES States now include Croatia (hence a total of 30), and the non‐SES States amount to 8 States. The data for SES States reflects latest available data on en‐route costs provided up to 15 January 2015.

Figure 6‐8 below presents the real en‐route unit costs calculated from these data, in €2009 and using the same metric as in RP2 (i.e. after deduction of costs for services to exempted VFR flights)37. Note that Georgia has joined the Multilateral Route Charged System as of 1 January 2014. However, Figure 6‐8 below does not include the data for Georgia, so as to have a consistent series from 2009 onwards.

Figure 6‐8: Pan‐European en‐route cost‐efficiency outlook 2015‐2019 (in €2009)

Figure 6‐8 shows that en‐route costs are foreseen to increase by +4.7% between 2013 and 2015 (i.e. on average by +2.3% p.a.) and then remain stable over the 2015‐2019 period. In the meantime, traffic is foreseen to increase by +7.1% between 2013 and 2015 (i.e. on average by +3.5% p.a.), followed by an average growth of +2.7% until 2019.

As a result, en‐route unit cost is expected to decrease from 53.1€2009 in 2013 to 46.6€2009 in 2019 (or ‐2.2% p.a. on average). Figure 6‐8 also indicates that at pan‐European level between 2009 and 2019, total en‐route costs are planned to remain flat, while traffic (SUs) is planned to increase by some +30%, implying substantial cost‐efficiency improvements over this 10‐years cycle.

37 This is different from the RP1 metrics (before deduction of costs for services to exempted VFR flights). Hence

the figure is not directly comparable to Figure 6‐2.

60.0

56.5

53.6 54.9

53.1 53.1 51.9

50.7

49.5 48.0 46.6

97 97 98

97

100 101 101 101 101 101

103

108 107

109

113

117 119

123

127

130

20

25

30

35

40

45

50

55

60

65

70

2009Actuals

2010Actuals

2011Actuals

2012Actuals

2013Actuals

2014Forecasts

2015Plan/

Forecast

2016Plan/

Forecast

2017Plan/

Forecast

2018Plan/

Forecast

2019Plan/

Forecast

En‐route real cost per SU (€2009)





77

6.3 Terminal ANS cost‐efficiency performance

The analysis of terminal ANS cost‐efficiency in this section refers to the RP1 SES States (see Figure 6‐9) which are required to provide terminal ANS costs and unit rates information in accordance with EU legislations [Ref. 26,27,28].

Although gradually improving, terminal ANS cost‐efficiency data have a lower level of maturity than en‐route ANS cost‐efficiency data. There is still some diversity in reporting between States and years which, to some extent, affects time series analyses and comparisons.

Terminal navigation charges are based on Terminal Navigation Service Units (TNSUs) which are computed as a function of the maximum take‐off weight ((MTOW/50)^α).

Up to 2015, the exponent (α) was not harmonised. However, in accordance with the Charging Scheme Regulation [Ref. 26], all States will have to use a common formula (MTOW/50)^0.7 as of 2015. In 2013, already 18 States had adopted this formula with France and Ireland following in 2014.

In order to ensure comparisons over time and between States in this section, the TNSUs between 2010 and 2013 for all States were computed by the EUROCONTROL Central Route Charges Office (CRCO) using the formula foreseen as of 2015. For 2014, the STATFOR 7‐year TNSU forecast (2014‐2020) published in February 2014 [Ref. 4] has been used38.

Terminal Navigation Charges vs. Airport Charges

Given the risk for potential misunderstanding, it is useful to differentiate between Terminal ANS charges (also called “TNC” for terminal navigation charges) and “Airport charges”, which typically include landing, passenger, cargo, parking and hangar, and noise charges, and are covered

by Directive 2009/12/EC [Ref. 29]. While

such airport charges amount to some €15 billion/year, the TNC in the SES represent some €1.5 billion/year.

Trends in actual terminal ANS cost‐efficiency performance

Figure 6‐10 provides a summary view of the actual terminal ANS cost‐effectiveness data between 2010 and 2013 and the forecast for 2014.

The number of reporting States/ Terminal Charging Zones (TCZ) increased between 2010 and 2012 by three States (Malta, Latvia, and Cyprus). As of 2012, a total of 29 States (31 terminal charging zones) reported on terminal ANS costs.

Figure 6‐10 shows that terminal ANS costs (1 354 M€2009) decreased by ‐3.9% in 2013, while at the same time traffic (TNSUs) decreased by ‐0.3% to 7.7 M TNSUs, leading to a ‐3.6% unit cost reduction (€175.4 per TNSU).

In 2014, real terminal ANS unit costs are forecast to increase by +3.5%. This is due to the fact that TNSUs are expected to slightly decrease (‐0.2%) while terminal ANS costs are foreseen to increase (+3.3%) between 2013 and 2014.

38 More details about the TNSU forecast method can be found in Methods of the STATFOR Seven‐Year Forecast.

Figure 6‐9: Geographical scope of terminal ANS cost‐efficiency analysis

SES states 2014


78

Figure 6‐10: Real terminal ANS unit costs (€2009) for reporting States

In absolute terms, the most significant decreases in terminal ANS costs were observed for:

Spain: ‐25.6M€2009 (‐16.1% vs. 2012), driven mainly by staff and operating costs savings, reflecting mainly the cost‐savings measures implemented by Aena; and,

Germany: ‐20.4M€2009 (‐9.1%), mainly due to significant reductions in staff and operating costs.

Other significant decreases of more than 10% compared to 2012: Luxembourg (‐23.1%), Sweden (‐14.8% for Arlanda TCZ and – 12.8% for Landvetter TCZ), Denmark (‐12.5%) and Greece (‐12.2%).

On the other hand, the largest increase in total costs in 2013 was observed in Norway (+7.8 M€2009, or +17.3%). Significant increases in percentage terms were also observed for: Slovakia (+23.6%), Romania (+17.0%) and Malta (+16.2%).

Terminal ANS cost‐efficiency analysis: 2013 unit costs by terminal charging zone

Figure 6‐11 shows the terminal ANS unit costs for the 31 TCZs in the 29 SES States (RP1) in 2013.

In 2013, terminal ANS costs per TNSU range from €621 for Slovak Republic TCZ to €67 for Sweden‐Landvetter TCZ, a factor of over nine. The two dotted lines in Figure 6‐11 represent the top and bottom quartiles of the dataset, giving an indication of the variance of calculated terminal ANS unit costs. In 2013, there were €89 per TNSU between the upper (€232) and lower (€143) quartiles, with the average of the proxy for the European unit cost amounting to €175.4 per TNSU39.

Slovakia TCZ’s high 2013 unit costs could be the result of relatively low traffic in relation to its total cost base. By comparison, Sweden‐Landvetter TCZ handled three times more traffic in 2013, at three times lower cost‐base than the Slovakian TCZ. As mentioned below, the scope of the Terminal ANS provided might be very different between the two TCZ.

39 It should be noted that the variation in unit cost between States shown in Figure 6‐11 does not vary substantially

if calculated using cost per movement instead of cost per TNSU.

198.8

185.8 182.0

175.4 181.6

98.0

94.6

90.9

93.9

104.8 103.3 103.0 102.8

120

130

140

150

160

170

180

190

200

210

2010Actuals

2011Actuals

2012Actuals

2013Actuals

2014Forecast

Term

inal real cost

per TNSU

(€2009)

Real terminal unit costs for the SES RP1 Area

(€2009)

TNSU index (2010)

Terminal ANS cost index (2010)


2010

Actuals

2011

Actuals

2012

Actuals

2013

Actuals

2014

Forecast

2013 vs

2012

2014F vs

2013

SES States reporting 26 28 29 29 29 0.0% 0.0%

Charging zones 28 30 31 31 31 0.0% 0.0%

Airports covered 224 226 229 229 229 0.0% 0.0%

1 489 1 459 1 409 1 354 1 398 ‐3.9% 3.3%

7.5 7.9 7.7 7.7 7.7 ‐0.3% ‐0.2%

198.8 185.8 182.0 175.4 181.6 ‐3.6% 3.5%

Total terminal ANS costs (M€2009)

Total terminal service units ((MTOW/50)^0.7, M TNSU)

Terminal real unit cost per TSNU (€2009)

Sources: State submissions to the European Commission for costs (November 2014 ), CRCO for actual TNSU, and STATFOR for forecast TNSU

Real terminal ANS unit costs for SES RP1 States

(€2009)

Number of


79

Figure 6‐11: Comparison of 2013 terminal ANS unit costs by TCZ (SES States (RP1))

Among the identified reasons for differences in terminal ANS unit cost are: the States’ discretion on defining their Terminal Charging Zones (TCZ), including the number of TCZ and the number and size of aerodromes; the charging policy, including charging formula applied until 2014 and applied cost‐allocation between en‐route and terminal; the traffic levels and complexity, and the scope of ANS provided. This introduces comparability issues when analysing and benchmarking terminal ANS performance levels across States/TCZ/airports.

Figure 6‐11 also shows that terminal ANS unit costs also substantially differ amongst the five largest States (from €242 for Italy TCZ to €89 for UK TCZ‐B).

Unit costs for terminal ANS looks particularly low in the UK TCZ B (€89 per TNSU). Firstly, it should be noted that the unit cost is not necessarily the same as the price charged for terminal ANS in the UK because of the contractual arrangements for the provision of terminal ANS. Low terminal ANS unit cost in UK TCZ B could be partly due to the fact that for the London airports (which account for most of the traffic in UK TCZ B), the cost data submitted only covers the aerodrome control service provided by NATS Services Ltd (NSL). In fact, Approach control for the London airports is provided by NATS En‐Route Ltd (NERL) and recovered through a separate London Approach Charge, for which no cost information is separately reported to the European Commission until 2014.

Another reason could be the significant larger scale of operations at the UK TCZ B (airports > 150 000 commercial movements) compared to any other TCZ. Finally, another explanation could be the greater cost‐efficiency provided by the UK model of potential “contestability” (now referred to as “market conditions”) for aerodrome ATC services. These particular issues would deserve further analysis and understanding to ensure a fair comparison and to identify genuine best practice performance management.

621

279

278

249

244

243

242

236

227

226

220

210

209

202

191

177

175

166

162

158

157

145

144

143

142

130

127

125

122

89

67

‐

50

100

150

200

250

300

350

400

450

500

550

600

650

Slovakia

Hungary

Luxembourg

Czech Rep

ublic

Switzerland

Slovenia

Italy

Bulgaria

Greece

Belgium

Norw

ay

Romania

France

Austria

Lithuania

Latvia

Cyprus

Ireland

Spain

Germany

UK‐Zone A

Den

mark

Poland

Netherlands

Portugal

Finland

Malta

Swed

en‐Arlanda

Estonia

UK‐Zone B

Sweden

‐LandvetterReal term

inal ANS cost per TNSU

(MOTW

/50)^0.7 in

€2009

European average (2013): €175.4

Upper quartile

Lower quartile


242 209 162 158 157 89 ‐

50

100

150

200

250

300Focus on the largest Terminal charging zones


80

Terminal ANS cost‐efficiency analysis: 2013 actuals versus 2013 forecasts

Figure 6‐12 below shows that terminal ANS costs in 2013 were ‐6.1% (or ‐87.8 M€2009) lower than forecasts used for the establishment of 2013 terminal ANS unit rates as reported in November 2012 by SES States.

Figure 6‐12: Comparison of 2013 terminal ANS actual costs vs. 2013 forecasts (Nov. 2012)

Note that a similar trend is also observed for en‐route, at system level there were no significant cost reallocation from en‐route towards terminal, and where the same ANSP provides both en‐route and terminal services, the cost‐efficiency improvement due to the SES target setting on en‐route is likely to also have had a positive impact on terminal ANS costs, mainly due to the level of shared/common costs.

As shown in Figure 6‐13 below, overall, actual 2013 terminal ANS costs were in 22 States lower than forecast in November 2012 and higher than forecast in 7 States. The largest reductions are observed for Germany (actual costs were lower ‐18.6 M€2009 than forecast provided in November 2012), Italy (‐16.1 M€2009), France (‐14.2 M€2009) and Spain (‐6.7 M€2009).

Figure 6‐13: 2013 Terminal ANS actual costs vs. 2013 forecast costs at State Level

Nov 2012 Reporting

Tables

Nov 2014

Reporting Tables

2013F 2013A

Real terminal ANS costs ‐ (in M€2009) 1 441.7 1 353.7 ‐88.0 ‐6.1%

SES States2013A vs

2013Fin %

4.8

0.9

0.6

0.4

0.4

0.3

0.3

0.1

‐0.2

‐0.2

‐0.3

‐0.4

‐1.0

‐1.2

‐1.3

‐1.4

‐1.6

‐1.9

‐2.0

‐2.3

‐2.6

‐2.8

‐3.0

‐3.3

‐4.4

‐4.7

‐5.7

‐6.7

‐14.2

‐16.1

‐18.6

‐25 M€

‐20 M€

‐15 M€

‐10 M€

‐5 M€

0 M€

5 M€

10 M€

Norw

ay

Slovakia

Malta

UK TCZ‐B

Lithuania

Bulgaria

Estonia

Cyprus

Swed

en‐Landvetter

Swed

en‐Arlanda

Austria

Romania

Slovenia

Ireland

Hungary

Finland

Greece

UK TCZ‐A

Portugal

Latvia

Czech Rep

ublic

Belgium

Den

mark

Netherlands

Switzerland

Poland

Luxembourg

Spain

France

Italy

Germany

Actual 2013 terminal ANS costs vs. Nov. 2012 forecasts (M€2009)



81

Figure 6‐14: Real terminal ANS costs per TNSU, total costs (€2009) and TNSUs

173.2

167.7 164.8

160.5 157.1

‐3.2%‐1.7%

‐2.6%‐2.1%

99.0 98.9 98.3 98.1

102.3 104.0106.1 108.2

120

130

140

150

160

170

180

2015F 2016F 2017F 2018F 2019FTerm

inal real cost per TNSU

(€2009)

Terminal SU index (2015)

Terminal ANS cost index (2015)

Source: NSAs; PRU analysis

Terminal ANS cost‐efficiency analysis: outlook for 2015‐2019

Figure 6‐14 shows that SES total terminal ANS costs are foreseen to slightly decrease (‐1.9%) over the period 2015‐2019 (i.e. on average by ‐0.5% p.a.) while TNSUs are foreseen to increase by some +8.2% (i.e. on average by +2.0% p.a.).

As a result, the forecast terminal ANS unit costs show a decrease from 173.2€2009 in 2015 to 157.1€2009 in 2019 (or ‐2.4% per year on average).

Figure 6‐15: Change in real terminal ANS total costs 2015‐2019 (real €2009)

Figure 6‐15 shows the planned change in real terminal ANS costs between 2015 and 2019 for all reporting States and TCZs. As discussed above, SES total terminal costs are expected to slightly decrease by ‐1.9% over the period. However, the figure above shows that some important changes in terminal costs are anticipated for some States/TCZs.

34%

15%

13%

11%

11%

10%

9%

9%

9%

9%

8%

6%

5%

3%

3%

3%

3%

3%

2%

2%

1%

0%

0%

0%

‐1%

‐1%

‐1%

‐2%

‐2%

‐3%

‐5%

‐6%

‐6%

‐8%

‐8%

‐11%

‐20%

‐10%

0%

10%

20%

30%

40%

Malta

Czech Republic

Greece

Portugal

Estonia

Luxembourg

Belgium EBAW

Belgium EBCI

Slovakia

Ireland

Romania

Belgium EBOS

Belgium EBLG

Belgium EBBR

United

Kingdom ‐ Zone C

Poland

Croatia

Cyprus

Austria

Italy Zone 2

Italy Zone 1

Finland

Netherlands

Switzerland

Latvia

Hungary

United

Kingdom ‐ Zone B

France

Lithuania

Swed

en Arlanda

Den

mark

Slovenia

Norw

ay

Bulgaria

Spain

Germany

Planned 2015‐2019 change in total terminal ANS

costs (real terms €2009)

Source: State submissions; PRU analysis


82

6.4 ANSPs gate‐to‐gate economic performance

The analysis of ANSPs economic performance in this section focuses on ATM/CNS provision costs i.e. those which are under the direct responsibility of the ANSP, plus the cost of delay attributable to ANSPs. Detailed analysis will be available in the ACE 2013 Benchmarking Report [Ref. 30].

The analysis developed in the ACE Reports allows identifying best practices in terms of ANSPs economic performance and to infer a potential scope for future performance improvements. This is a useful complement to the analysis of the en‐route KPI and terminal PIs which are provided in the previous sections of this chapter.

Figure 6‐16 shows a detailed breakdown of gate‐to‐gate ATM/CNS provision costs. Since there are differences in cost‐allocation between en‐route and terminal ANS among ANSPs, it is important to keep a “gate‐to‐gate” perspective when benchmarking ANSPs cost‐effectiveness performance.

Figure 6‐16: Breakdown of gate‐to‐gate ATM/CNS provision costs 2013 (€2013)

Figure 6‐16 indicates that in 2013, at European system level, gate‐to‐gate ATM/CNS provision costs amount to some €8.0 Billion. Operating costs (including staff costs, non‐staff operating costs and exceptional cost items) account for some 82% of total ATM/CNS provision costs, and capital‐related costs (cost of capital and depreciation) amount to some 18%.

The analysis presented in this section is factual. It is important to note that local performance is affected by several factors which are different across European States, and some of these are typically outside (exogenous) an ANSP’s direct control while others are endogenous. Indeed, ANSPs provide ANS in contexts that differ significantly from country to country in terms of environmental characteristics (e.g. the size and complexity of the airspace), institutional characteristics (e.g. relevant State laws), and of course in terms of operations and processes.


83

A genuine measurement of cost inefficiencies would require full account to be taken of the exogenous factors which affect ANSPs economic performance. This is not straightforward since these factors are not all fully identified and measurable. Exogenous factors related to operational conditions are, for the time being, those which have received greatest attention and focus. Several of these factors, such as traffic complexity and seasonal variability, are now measured robustly by metrics developed by the PRU.

The quality of service provided by ANSPs has an impact on the efficiency of aircraft operations, which carry with them additional costs that need to be taken into consideration for a full economic assessment of ANSP performance. The quality of service associated with ATM/CNS provision by ANSPs is, for the time being, assessed only in terms of ATFM ground delays, which can be measured consistently across ANSPs, can be attributed to ANSPs, and can be expressed in monetary terms. The indicator of “economic” cost‐effectiveness is therefore the ATM/CNS provision costs plus the costs of ATFM ground delay, all expressed per composite flight‐hour.

A number of factors affecting aircraft operations and contributing to the quality of service that is provided to airspace users by an ANSP are not accounted for in the economic cost‐effectiveness indicator analysed in this report. These include operational aspects such as:

horizontal flight‐efficiency and the resulting route length extension; and,

vertical flight‐efficiency and the resulting deviation from optimal vertical flight profile.

There is no mature and commonly agreed methodology to measure the horizontal and vertical flight‐efficiency genuine contribution at ANSP level. Therefore the ACE Benchmarking Report continues to focus on the costs of gate‐to‐gate ATFM ground delays to benchmark ANSPs cost‐effectiveness. The analytical framework is illustrated in Figure 6‐17.

Figure 6‐17: Conceptual framework for the analysis of economic cost‐effectiveness

Trends in economic cost‐effectiveness (2009‐2013)

Figure 6‐18 below displays the trend at European level of the gate‐to‐gate economic costs per composite flight‐hour (“unit economic costs” thereafter) between 2009 and 2013 for a consistent sample of 37 ANSPs for which data for a time‐series analysis was available. In 2009, the economic recession affected the aviation industry with an unprecedented ‐7% traffic decrease at system level, basically cancelling three years of traffic growth. It is therefore interesting to look at the changes in performance over the 2009‐2013 period to understand how the ATM industry reacted to this sharp decrease in traffic demand.

Costs of ATFM delays

Economic cost‐effectiveness

indicator(unit economic costs)

EUROCONTROL/PRU

Composite flight‐hours

ATFM delay costs per unit of

output

Financial cost‐effectiveness

indicator(unit ATM/CNS provision costs)

Inputs

PerformanceIndicators

Outputs

ATM/CNS

provision costs

IFR airport movements

En‐route flight‐hours

Composite flight‐hours

The "composite gate‐to‐gate flight‐hours" combines the two separate output measures for en‐route (i.e. flight‐hours) and terminal ANS (i.e. airport movements). Composite flight‐hours are computed by weighting the en‐route and terminal output measures using their respective unit costs. This average weighting factor is calculated at European system level using ANSPs costs and outputs data relating to the period 2002‐2013 and amounts to 0.27.

The composite flight‐hours are therefore defined as:

En‐route flight‐hours + (0.27 × airport movements)

Although the composite gate‐to‐gate output metric does not fully reflect all aspects of the complexity of the services provided, it is nevertheless the best metric currently available for the comparison of gate‐to‐gate ANSP economic performance.


84

Figure 6‐18: Changes in economic cost‐effectiveness, 2009‐2013 (€2013)

At system level, unit economic costs increased in 2010 (i.e. +5.2% in real terms) and then reduced over 2011‐2013 (i.e. ‐4.2% p.a. in real terms) mainly due to the significant decreases in ATFM delays unit costs (‐32.3% p.a. over the 2010‐2013 period).

The right‐hand side of Figure 6‐18 indicates that in 2013, ATM/CNS provision costs reduced by ‐2.0% while the number of composite flight‐hours remained fairly constant. As a result, unit ATM/CNS provision costs decreased by ‐1.9% in 2013. In the meantime, the unit costs of ATFM delays significantly fell (‐18.2%) contributing to the decrease in unit economic costs observed in 2013 (‐3.6%). Unit economic costs amounted to €478 in 2013, which is ‐13.0% lower than pre‐recession levels (€549 in 2008).

Figure 6‐19 below shows that between 2012 and 2013, unit economic costs fell for 21 ANSPs. For six of these ANSPs, the main driver for the decrease in unit economic costs was a reduction in ATFM delays. This is particularly the case for DFS, DHMI and Croatia Control.

Figure 6‐19: Changes in economic cost‐effectiveness by ANSP, 2012‐2013 (€2013)

It is important to note that, for ANSPs operating outside of the Euro zone, substantial changes of the national currency against the Euro may significantly affect the level of 2013 unit ATM/CNS provision costs when expressed in Euro. Significant changes are observed over the 2003‐2013 period for several ANSPs part of the ACE analysis. For example, the Swiss Franc significantly appreciated (24%) while the British Pound substantially depreciated (18%). For most of these ANSPs, this mainly reflects substantial variations in exchange rates compared to the Euro occurred during the 2008‐2012 timeframe in the wake of the financial crisis and economic downturn.

For most of the ANSPs, the main driver for the reduction in unit economic costs in 2013 was a decrease in unit ATM/CNS provision costs (see blue bar in Figure 6‐19). Figure 6‐22 provides a

€552€581

€521 €496 €478

0

100

200

300

400

500

600

700

2009 2010 2011 2012 2013

€per composite flight‐hour (2013

prices)

ATFM Delay costs per composite flight‐hour

ATM/CNS provision costs per composite flight‐hour

‐4.3%

+1.8%

‐0.2%‐2.0%

+2.1% +3.9%

‐1.9%‐0.1%

‐37.6% ‐39.3%

‐18.2%

‐40%

‐30%

‐20%

‐10%

0%

10%

20%

30%

40%

2009‐10 2010‐11 2011‐12 2012‐13

ATM/CNS provision costs Composite flight‐hours Unit costs of ATFM delays

+77.4%

0

100

200

300

400

500

600

700

800

900

Belgocontrol

Skyguide

LVNL

LPS

DCAC Cyprus

DFS

Austro Control

Slovenia Control

Aen

a

UkSATSE

DSN

A

ENAV

NATS (Continen

tal)

ROMATSA

M‐NAV

ARMATS

MoldATSA

ANS CR

Albcontrol

Avinor (Continental)

HungaroControl

Croatia Control

Oro Navigacija

Finavia

LFV

NAVIAIR

PANSA

BULATSA

NAV Portugal (Continen

tal)

HCAA

IAA

SMATSA

DHMI

MUAC

LGS

EANS

MATS

€per composite flight‐hour (2013 prices)

ATM/CNS provision costs per composite flight‐hour (2013)

ATFM delays costs per composite flight‐hour (2013)

Unit economic costs (2012)


85

detailed analysis of the changes in unit ATM/CNS provision costs at ANSP level between 2012 and 2013, separately identifying:

Changes in ATM/CNS provision costs (the “cost effect”);

Changes in composite flight‐hours (the “traffic effect”).

Figure 6‐20: Changes in ATM/CNS provision costs and traffic, 2012‐2013 (€2013)

Figure 6‐22 shows that 18 ANSPs could reduce their ATM/CNS provision costs between 2012 and 2013 (see lower part of the chart). Out of the five largest ANSPs, Aena (‐11.3%) and DFS (‐4.9%) could achieve a reduction in ATM/CNS provision costs in a context of traffic decrease (‐5.7% and ‐2.6%, respectively). On the other hand, in 2013 ATM/CNS provision costs increased for NATS (+6.4%) and remained fairly constant for DSNA (+0.3%) and ENAV (+0.03%).

For NATS, the increase in ATM/CNS provision costs (+6.4%) is mainly associated to higher exceptional costs (+€44.2M) while traffic remained fairly constant (+0.4%). These higher exceptional costs mainly reflect a voluntary redundancy programme which will lead to the departure of some 240 employees.

Further details about the changes in ATM/CNS provision costs between 2012 and 2013 at ANSP level can be found in the ACE 2013 Benchmarking Report.

Breakdown of unit ATM/CNS provision costs (2009‐2013)

Figure 6‐21 shows how the unit ATM/CNS provision costs can be broken down into three main key economic drivers: (1) ATCO‐hour productivity, (2) employment costs per ATCO‐hour and (3) support costs per composite flight‐hour. Note that support costs represent 70% of total ATM/CNS provision costs and include:

employment costs for non‐ATCO in OPS staff; these cover ATCOs on other duties, trainees, technical support and administrative staff (48.7% of support costs);

non‐staff operating costs mostly comprise expenses for energy, communications, contracted services, rentals, insurance, and taxes (23.7% of support costs);

exceptional costs (2.0% of support costs); and,

capital‐related costs, comprising depreciation and financing costs for the capital employed (25.6% of support costs).

DSNA

DFS

Aena

NATS (Continental)

ENAV

DHMI

Skyguide

UkSATSE

Avinor (Continental)

LFV

Austro Control

LVNL

ROMATSA

Belgocontrol

HCAAPANSA

MUAC

NAV Portugal (Continental)

ANS CR

NAVIAIR

IAA

HungaroControl

Croatia Control

BULATSA

SMATSA

Finavia

LPS

DCAC Cyprus

Slovenia Control

Oro Navigacija

LGS

Albcontrol

EANS

MATS

MoldATSA

M‐NAVARMATS

‐20%

‐15%

‐10%

‐5%

0%

5%

10%

15%

20%

‐10% ‐5% 0% 5% 10% 15%

Changes in ATM

/CNS provision costs (2012‐2013)

Changes in composite flight‐hours (2012‐2013)

Increase in unit ATM/CNS provision costs

Decrease in unit ATM/CNS provision costs


86

Gate‐to‐gate ATM/CNS provision costs per composite flight‐hour

Gate‐to‐gate ATCO‐hour productivity

Gate‐to‐gate employment costs per ATCO‐hour Gate‐to‐gate support costs per composite flight‐hour

Figure 6‐21: ATM/CNS cost‐effectiveness comparisons, 2009‐2013 (€2013)

Figure 6‐22 shows how the various components contributed to the overall change in cost‐effectiveness between 2012 and 2013.

In 2013, employment costs per ATCO‐hour rose faster (+1.4%) than ATCO‐hour productivity (+0.9%), resulting in slightly higher ATCO employment costs per composite flight‐hour (+0.5%) compared to 2012. Figure 6‐22 also shows that support costs decreased by ‐3.0% while traffic volumes remained fairly constant compared to 2012, and as a result support costs per composite flight‐hour reduced by ‐2.9%. The central part of Figure 6‐22 shows that between 2012 and 2013, given the respective weights of ATCO employment costs (30%) and support costs (70%), unit ATM/CNS provision costs fell by ‐1.9%.

Figure 6‐22: Breakdown of changes in cost‐effectiveness, 2012‐2013 (€2013)

Further details about the changes in unit ATM/CNS provision costs, ATCO‐hour productivity, employment costs per ATCO‐hour and support costs at ANSP level can be found in the ACE 2013 Benchmarking Report.

477447 438 445 436

0

100

200

300

400

500

2009 2010 2011 2012 2013


prices)

0.73 0.78 0.80 0.80 0.81

0.0

0.2

0.4

0.6

0.8

1.0

2009 2010 2011 2012 2013

Composite flight‐hour per ATCO‐hour

on duty

€108€103 €105 €107 €108

0

20

40

60

80

100

120

2009 2010 2011 2012 2013

€per ATCO‐hour on duty (2013 prices)

€327€314 €305 €311 €302

0

50

100

150

200

250

300

350

2009 2010 2011 2012 2013


prices)

+0.9%+1.4%

+0.5%

-1.9%

-2.9% -3.0%

-0.1%

"Traffic effect"

Increased ATCO-hour productivity

Decrease inunit ATM/CNS

provision costs2012-2013

"Support costs effect"

Increased employment

costs per ATCO-hour

Increased ATCO employment

costs per composite flight-hour

Decreased support costs per composite

flight-hour

Weight 70%

Weight 30%


87

6.5 Conclusions

PRR2014 analyses performance in 2014 for all KPAs, except for cost‐efficiency, which analyses performance in 2013 as this is the latest year for which actual financial data are available.

The pan‐European system en route cost‐efficiency performance improved in 2013. Real en‐route costs per service unit decreased by ‐3.3% compared to 2012, reflecting a decrease of ‐1.3% in total costs and an increase of +2.1% in SUs. Total en‐route costs were also ‐5.6% lower than planned, responding to SUs which were ‐5.3% lower than forecast.

However, costs can be significantly impacted year on year by changes in (accounting) provisions for future liabilities (mainly for pensions). As highlighted in PRR 2013, the volatility of these (accounting) provisions raises concerns in the context of charging and performance, as changes in these provisions, which can be significant, do not necessarily represent costs directly attributable to the provision of ANS in the year in which they are recorded.

For the SES States, 2013 is the second year of application of the “determined costs” method with specific risk‐sharing arrangements. For the other nine EUROCONTROL States participating in the Route Charges System, the “full cost‐recovery method” continued to apply in 2013. The evaluation between SES states and the 9 other states does not yet show clear trends.

Under the determined costs method, the amounts ultimately paid by the airspace users differ from the actual costs due to the traffic risk‐sharing, cost‐sharing and other adjustments provided in the Charging Regulation. The PRC computes that, in 2013, these “true costs for users” are +3.5% higher than the actual costs of States/ANSPs but ‐2.0% lower than the determined costs provided for 2013 in the RP1 performance plans. This should be seen in the context of actual costs in 2013 which are ‐5.9% lower than planned in the RP1 performance plans. The fact that several States/ANSPs demonstrated lower costs than planned was duly considered during the RP2 assessment process in order to ensure that users benefit from the cost reductions achieved during RP1.

The outlook for 2014‐2019 suggests that the recent improved cost performance will not continue. Instead current plans/forecasts show costs stabilizing over the period. The forecast decrease in en‐route unit cost (‐2.2% per year on average between 2013 and 2019) is due to the forecast increase in SUs.

Terminal ANS unit costs (TNSUs) decreased by ‐3.6% between 2012 and 2013 reflecting a reduction of ‐3.9% in terminal ANS costs in real terms and a small traffic decrease (‐0.3%, TNSU).

The outlook for 2015‐19 shows a slower decline in costs. Terminal ANS unit costs are forecast to decrease by ‐2.4% per year on average, reflecting an average reduction of ‐0.5% p.a. in costs and an average increase of +2.0% p.a. in TNSUs

Gate‐to‐gate unit economic costs (i.e. en‐route plus terminal ATM/CNS provision costs plus taking account of the cost of delay) decreased by ‐3.6% in 2013, reflecting both the reduction in ATFM delays unit costs compared to 2012 (‐18.2%) and a decrease in gate‐to‐gate unit ATM/CNS provision costs (‐1.9%). However the increased levels of ATFM delays observed in 2014 suggests that unit economic costs could be higher in 2014.

The benchmarking analysis of gate‐to‐gate unit ATM/CNS provision costs indicates that the reduction of ‐1.9% in 2013 compared to 2012 is mainly due to the reduction of ‐3.0% in support costs (70% of total costs). Over the 2009‐2013 period, pan‐European unit ATM/CNS provision costs significantly reduced by ‐2.2% p.a., reflecting a gradual decrease in unit support costs and increasing ATCO productivity while ATCO employment costs remained fairly constant.

88

ANNEX I - IMPLEMENTATION OF PC DECISIONS

Implementation status of PC decisions relating to the PRC recommendations contained in PRR 2013 (published May 2014)

PC 41 decision (May 2014) on PRC’s recommendations in PRR

2013

PRC’s rationale for the recommendation

STATUS

r‐1. The Provisional Council encouraged States to review their State Safety Programmes and Safety Plans to ensure that major risks are being addressed.

Although the accident statistics show a reduction in ANS related accidents in recent years, serious incident statistics demonstrate that mid‐air collision and runway incursion accidents could still occur.

EASA monitors the implementation of the State Safety Programmes and Safety plans.

r‐2. The Provisional Council urged those States that have not yet done so to implement the Provisional Council’s Decisions 8.1b and c (PC 39, May 2013) as a matter of urgency.

The 2011 PRC recommendations requesting improvement in safety data reporting and safety data quality are not yet adequately implemented and therefore are reiterated as last year.

A specific task force was set up to address the issues. Preliminary 2014 data was received by a record number of 40 EUROCONTROL Member States through the AST mechanism.

r‐3. The Provisional Council requested States to ensure that sufficient resources are made available for: (i) the reporting, investigation,

storing, and analysis of ATM safety occurrences; and,

(ii) the risk assessment and hence severity classification of all reported ATM safety occurrences.

The number of un‐assessed incidents is increasing since 2007. This situation is of concern, not only for the outcome of the analysis at European level, but also for the national safety analysis.

The issue is addressed again in Recommendation b of PRR 2014.

r‐4. The Provisional Council urged States to support the inclusion of specific provisions regarding the severity classification of ATM occurrences in their safety regulatory framework.

Where no existing regulations are in place, States should support the inclusion of specific provisions regarding the severity classification of ATM occurrences in their safety regulatory framework.

The progress is monitored by the PRC.

Ongoing

r‐5. The Provisional Council encouraged States to expedite the deployment of automatic safety data reporting/monitoring tools in Europe to improve the identification of safety risks and to measure the effectiveness of safety improvement action.

There is an urgent need to accelerate the deployment of automatic safety data reporting tools to improve the identification of safety risks and measure the effectiveness of safety improvement action. Deployment of such tools should also improve the reporting culture and consequently the level of reporting.

The implementation process is monitored via the LSSIP.

r‐6. The Provisional Council requested the Director General to initiate a pan‐European harmonisation project, in order to use data from automated safety data reporting/monitoring tools at State, FAB or European level, because the event‐triggers for each type of occurrence need to be harmonised.

There is a need to conduct a pan‐European harmonisation project to ensure that data can be shared and aggregated.

A letter from the Director General EUROCONTROL dated 10.10.2014 supports this recommendation.

Follow‐up discussions between the PRC and the Network Manager are ongoing.

89


2013


STATUS

r‐7. The Provisional Council urged the States and ANSPs to ensure that updated information on airspace and route availability is transmitted to the Network Manager in a timely manner and made available to airspace users for flight planning purposes.

The analysis reveals a considerable gap between flight plan and actual flight trajectories with an impact on fuel burn (additional carriage) and capacity (lower level of predictability of operations).

To reduce the gap there is a need to ensure that the Network Manager and airspace users are continuously supplied with the latest information on airspace and route availability.

The progress is monitored by the PRC and a briefing by the NM was provided at PRC 88 in April 2015.

Ongoing

r‐8. The Provisional Council requested the Director General to initiate the development of tools to identify the shortest route available at the time the flight commences.

Due to the lack of information on route availability, it is presently not possible to identify the shortest available route at the time the flight commences.

DG letter dated 10.10.2014 supports this recommendation.

Follow‐up discussions between the PRC and NM are ongoing.

r‐9. The Provisional Council urged States to ensure an accurate and consistent classification of ATFM delays to enable constraints on European ATM to be correctly identified and resolved or mitigated.

Investigation into the specific classification of ATFM delay highlighted inconsistency in how delays are assigned both in terms of the causal factor and the appropriate location. Such inconsistency is detrimental to performance improvement.

While issues still remain in some States, NM has developed a process for the post operational review of ATFM delay allocation and classification, whenever issues have been identified by stakeholders.

r‐10. The Provisional Council urged the Director General to work with non‐ECAC States to improve the predictability of traffic entering the EUROCONTROL area, in order to improve the service being provided to air traffic within the EUROCONTROL area.

“Out of area” flights show a lower level of predictability with an impact on safety, service quality and capacity utilisation in the periphery of the EUROCONTROL area. An enhanced exchange of information would help improving the situation.

NM has been engaging with various States to improve the predictability of traffic entering the EUROCONTROL area.

Ongoing

r‐11. The Provisional Council encouraged States to fully implement Airport Collaborative Decision Making (A‐CDM) including Departure Planning Information (DPI) exchange with Network Operations at congested airports, in line with the related European Single Sky Implementation Plan and the NM Performance Plan.

A‐CDM is an enabler for improving the efficiency of the departure flow management at airports and also ensures a better integration of airports into the ATM network which improves network predictability and the utilisation of available resources.

The PRC is closely following the A‐CDM implementation in Europe.

In 2014, Rome Fiumicino and London Gatwick became fully implemented A‐CDM airports.

r‐12. The Provisional Council recalled its Decisions g) and f) at PC 39 (May 2013) encouraging those States not bound by the provisions of the SES performance scheme to provide Correlated Position Reports and data on operations at key airports (>70 000 movements), using the standards described in the SES legislation, to enable EUROCONTROL‐wide data collection and ANS performance review to be harmonised.

Presently data provision in EUROCONTROL States is not harmonised. The recommendation aims at achieving a harmonised CPR (30 seconds radar update rate) and airport data collection following the standards laid out in the SES legislation. The claim is supported by sections in the en route chapter (flight efficiency) and in the airport chapter.

The progress is monitored by the PRC and a briefing by the NM was provided at PRC 88 in April 2015.

Ongoing

90


2013


STATUS

r‐13. The Provisional Council encouraged those States not bound by the provisions of the SES performance scheme that wish to apply the “determined costs” method to provide for independent supervision by their national authority and a meaningful consultation with airspace users and to take due account of the level and ambition of performance improvements expected for SES States.

A recent revision (November 2013) of the “EUROCONTROL Principles for establishing the cost‐base for en route charges and the calculation of the unit rates” gives the possibility for the States which are not bound by the SES to opt for either the “full cost‐recovery method” or the “determined costs method”. It can therefore be expected that some non‐SES States will apply the “determined costs method” in the future, given the incentive possibilities offered by this method. The supervision by their national authority and assessment of the level of the “determined costs” and associated unit rates charged to users is an integral part of the “determined costs method”.

States not bound by the provisions of the SES performance scheme that wish to apply the “Determined Costs” method are invited to follow this recommendation.

r‐14. The Provisional Council urged States to take due account of the fairness, cost‐relatedness and appropriateness of a strict application of International Financial Reporting Standards when establishing air navigation cost‐bases and charges

The volatility of the (accounting) provisions raises concerns in the context of charging and performance, whereas increases or decreases in these provisions do not necessarily represent costs directly attributable to the provision of ANS in the year in which they are recorded. Moreover, these increases or decreases in provisions, especially when related to pensions, can be significant in size and thereby influence significantly the resulting cost‐efficiency indicator, which may no longer reflect the adjustment of costs to the traffic context and the genuine cost‐efficiency performance of States/ANSPs or even the Pan‐European system as a whole.

States are invited to follow this recommendation when establishing their air navigation cost‐bases and charges.

91

ANNEX II - ACC TRAFFIC AND DELAY DATA (2012-2014)

Source: NM

Please note that delay per flight is not an additive measure (on average a flight crosses between 2 and 3 States). Therefore, the European value does not transpose directly to individual ACCs.

3Y‐AAGR = Annual average growth rate

State ACC

2012

2013

2014

2012

2013

2014

2012

2013

2014 Capacity/

Staffing

ATC

OtherWeather

Other

reasons

Albania Ti rana 533 550 543 ‐1.2% 0.1% 0.06 0.00 0.00 0.06 0.00 0.00

Armenia Yerevan 144 139 130 ‐6.1% ‐3.8% 0.00 0.00 0.00 0.00 0.00 0.00

Austria Wien 1 961 1 916 2 057 7.3% 0.7% 0.38 0.48 0.17 0.16 0.26 0.03 12.5% 4.3% 83.2% 0.0%

Belgium Brussels 1 503 1 483 1 525 2.9% ‐0.5% 0.17 0.26 0.21 0.03 0.08 0.02 47.4% 33.8% 6.8% 12.0%

Bulgaria Sofia 1 422 1 460 1 822 24.8% 8.7% 0.01 0.00 0.00 0.00 0.00 0.00

Croatia Zagreb 1 286 1 281 1 355 5.8% 1.7% 0.28 0.10 0.33 0.27 0.10 0.33 50.2% 9.4% 38.6% 1.8%

Cyprus Nicos ia 736 760 834 9.7% 2.7% 1.65 2.21 1.92 1.59 2.16 1.91 68.0% 3.2% 1.2% 27.7%

Czech Republ ic Praha 1 793 1 804 1 849 2.5% 0.1% 0.01 0.06 0.03 0.00 0.04 0.01 65.5% 0.0% 22.5% 12.0%

Denmark Kobenhavn 1 409 1 459 1 464 0.3% ‐0.3% 0.02 0.02 0.01 0.00 0.00 0.00 0.0% 0.0% 100.0% 0.0%

Estonia Tal l inn 493 485 508 4.7% 2.7% 0.11 0.02 0.03 0.11 0.02 0.03 90.9% 0.0% 8.7% 0.5%

Finland Tampere+ 485 451 459 1.8% ‐4.9% 0.26 0.03 0.26 0.02 0.00 0.16 0.0% 0.0% 0.0% 100.0%

France Bordeaux 2 222 2 238 2 282 2.0% 0.6% 0.30 0.33 0.24 0.27 0.30 0.23 22.0% 39.8% 7.7% 30.5%

Brest 2 397 2 457 2 559 4.2% 1.6% 0.20 0.36 0.54 0.19 0.35 0.53 46.3% 48.1% 1.8% 3.8%

Marsei l le 2 763 2 746 2 730 ‐0.6% ‐0.9% 0.55 0.44 0.57 0.55 0.44 0.57 33.3% 59.0% 5.9% 1.9%

Paris 3 227 3 106 3 125 0.6% ‐1.6% 0.58 0.42 0.35 0.27 0.17 0.17 47.1% 21.8% 24.8% 6.4%

Reims 2 334 2 430 2 522 3.8% 3.0% 0.27 0.33 0.43 0.26 0.31 0.42 78.2% 7.2% 11.9% 2.7%

FYROM Skopje 306 301 389 29.5% 4.6% 0.00 0.00 0.00 0.00 0.00 0.00 100.0% 0.0% 0.0% 0.0%

Germany Bremen 1 674 1 628 1 683 3.4% ‐0.5% 0.16 0.16 0.18 0.06 0.06 0.09 48.6% 14.1% 32.8% 4.5%

Langen 3 376 3 318 3 317 0.0% ‐1.1% 1.03 0.46 0.53 0.64 0.24 0.24 44.6% 0.6% 33.0% 21.8%

Munchen ++ 3 911 2 876 2 846 ‐1.1% ‐11.3% 0.49 0.14 0.12 0.32 0.05 0.02 4.2% 1.7% 93.5% 0.6%

Rhein (Karl sruhe) 3 905 4 501 4 631 2.9% 6.2% 0.09 0.17 0.20 0.09 0.17 0.20 32.3% 2.8% 53.7% 11.3%

Greece Athina i+Macedonia 1 673 1 643 1 782 8.4% 0.8% 0.34 0.32 0.78 0.16 0.07 0.42 94.3% 3.9% 1.7% 0.2%

Hungary Budapest 1 526 1 566 1 754 12.0% 3.2% 0.00 0.00 0.00 0.00 0.00 0.00 64.8% 35.2% 0.0% 0.0%

Ireland Dubl in 491 509 537 5.4% 3.2% 0.05 0.05 0.02 0.00 0.00 0.00

Shannon 1 075 1 074 1 086 1.1% ‐0.1% 0.00 0.00 0.00 0.00 0.00 0.00

Ita ly Brindis i 808 786 730 ‐7.1% ‐5.7% 0.02 0.05 0.01 0.00 0.00 0.01 0.0% 100.0% 0.0% 0.0%

Mi lano 1 659 1 567 1 973 25.9% 4.7% 0.02 0.01 0.03 0.00 0.00 0.03 0.0% 100.0% 0.0% 0.0%

Padova 1 844 1 821 1 854 1.8% ‐0.2% 0.03 0.02 0.02 0.00 0.00 0.01 0.0% 100.0% 0.0% 0.0%

Roma 2 583 2 564 2 239 ‐12.7% ‐5.6% 0.07 0.09 0.12 0.00 0.00 0.00 1.1% 98.9% 0.0% 0.0%

Latvia Riga 634 642 659 2.6% 1.0% 0.00 0.00 0.00 0.00 0.00 0.00

Lithuania Vi lnius 544 565 597 5.6% 3.8% 0.00 0.00 0.00 0.00 0.00 0.00

Maastricht 4 387 4 471 4 579 2.4% 1.3% 0.04 0.07 0.17 0.04 0.07 0.17 49.2% 0.0% 26.3% 24.5%

Malta Malta 264 298 277 ‐7.0% 7.7% 0.00 0.00 0.01 0.00 0.00 0.00

Moldova Chis inau 171 198 149 ‐24.8% ‐2.9% 0.00 0.00 0.00 0.00 0.00 0.00

The Netherlands Amsterdam 1 393 1 408 1 441 2.3% 0.6% 0.78 0.68 0.94 0.18 0.12 0.13 70.4% 0.0% 23.5% 6.1%

Norway Bodo 555 565 589 4.1% 2.7% 0.03 0.03 0.03 0.03 0.03 0.02 12.3% 87.7% 0.0% 0.0%

Os lo 898 949 961 1.2% 2.8% 0.96 0.38 0.29 0.47 0.00 0.01 100.0% 0.0% 0.0% 0.0%

Stavanger 625 663 677 2.0% 4.8% 0.03 0.12 0.24 0.01 0.07 0.05 94.8% 0.0% 0.0% 5.2%

Poland * Warszawa 1 723 1 753 1 775 1.2% 1.8% 0.56 0.56 0.88 0.56 0.54 0.84 67.7% 0.5% 8.2% 23.6%

Portugal Lisboa 1 121 1 150 1 229 6.9% 2.1% 0.95 0.43 0.72 0.69 0.29 0.53 96.2% 1.8% 1.9% 0.2%

Santa Maria 295 305 315 3.3% 0.9% 0.00 0.00 0.00 0.00 0.00 0.00

Romania Bucuresti 1 308 1 383 1 617 16.9% 6.6% 0.00 0.00 0.00 0.00 0.00 0.00

Serbia Beograd 1 435 1 393 1 491 7.0% ‐0.2% 0.00 0.02 0.01 0.00 0.02 0.00 48.1% 0.4% 10.8% 40.7%

Slovak Republ ic Brati s lava 2 050 2 215 2 476 11.7% 7.3% 0.63 0.15 0.68 0.00 0.00 0.00

Slovenia Ljubjana 735 703 743 5.7% 0.1% 0.00 0.00 0.00 0.00 0.00 0.00 100.0% 0.0% 0.0% 0.0%

Spain Barcelona 2 013 2 007 2 042 1.8% ‐1.5% 0.68 0.49 0.46 0.63 0.47 0.37 72.8% 0.0% 26.8% 0.3%

Madrid 2 500 2 395 2 514 5.0% ‐2.7% 0.34 0.22 0.10 0.18 0.18 0.07 46.0% 11.3% 2.1% 40.6%

Palma 682 674 695 3.1% ‐1.0% 0.57 0.48 0.49 0.19 0.13 0.11 100.0% 0.0% 0.0% 0.0%

Sevi l la 894 879 901 2.5% ‐3.5% 0.09 0.07 0.03 0.06 0.05 0.03 74.5% 0.0% 19.8% 5.7%

Canarias 749 724 774 6.9% ‐1.6% 0.57 0.56 0.44 0.38 0.44 0.42 75.6% 0.4% 22.8% 1.1%

Sweden Malmo 1 359 1 377 1 386 0.6% ‐0.1% 0.04 0.00 0.01 0.04 0.00 0.01 8.9% 48.2% 43.0% 0.0%

Stockholm 1 062 1 069 1 078 0.8% ‐0.5% 0.15 0.17 0.16 0.02 0.05 0.05 2.6% 57.5% 32.4% 7.6%

Switzerland Geneva 1 654 1 627 1 654 1.7% ‐1.0% 0.26 0.41 0.34 0.06 0.10 0.10 73.1% 4.8% 21.8% 0.4%

Zurich 2 031 1 975 1 984 0.5% ‐1.5% 0.62 0.61 0.56 0.19 0.14 0.08 77.5% 5.4% 16.7% 0.4%

Turkey Ankara 1 928 2 037 2 302 13.0% 6.3% 0.23 0.19 0.15 0.21 0.14 0.12 32.4% 0.1% 0.0% 67.6%

Is tanbul 2 050 2 215 2 476 11.7% 7.3% 0.63 0.15 0.68 0.00 0.00 0.00

Ukra ine Kyiv 631 651 532 ‐18.2% ‐4.3% 0.04 0.02 0.01 0.00 0.00 0.00

Dnipropetrovs 'k ALL** 427 447 226 ‐49.5% ‐17.6% 0.00 0.00 0.00 0.00 0.00 0.00

Simferopol 540 594 134 ‐77.5% ‐37.4% 0.00 0.00 0.00 0.00 0.00 0.00

L'viv 485 502 363 ‐27.7% ‐9.0% 0.00 0.00 0.00 0.00 0.00 0.00

Odesa 268 299 285 ‐4.7% 3.1% 0.00 0.00 0.00 0.00 0.00 0.00

United Kingdom London AC 4 894 4 927 5 033 2.2% 0.4% 0.07 0.14 0.05 0.07 0.14 0.05 49.8% 3.5% 20.7% 26.0%

London TC 3 386 3 408 3 511 3.0% 0.9% 0.65 0.60 0.48 0.02 0.01 0.01 29.7% 0.0% 68.7% 1.6%

Prestwick 2 380 2 397 2 400 0.2% ‐0.7% 0.07 0.04 0.04 0.02 0.00 0.02 93.7% 0.0% 6.3% 0.0%

ACCs geographical areas might change over time, preventing year on year comparision (e.g. Prestwick, Dnipropetrovs'k ALL)

* does not include EPWWICTA and EPKKTMA

** Dnipropetrovs'k ALL was created in March 2010 replacing Kharkiv, Dnipropetrov'k and Donetsk' ACCs

+ Rovaniemi ACC was merged with Tampere ACC in 2011. ++ Upper airspace was transferred to Karlsruhe in Aug. 2012.

Causes of en route

ATFM delay in 2014En route ATFM

delay per fl ight

IFR Traffi c

2014/13

growth

(%)

3Y‐

AAGR

Avg. dai ly

Tota l ATFM

delay per fl ight

92

ANNEX III - TRAFFIC COMPLEXITY SCORES IN 2014

The PRU, in close collaboration with ANSPs, has defined a set of complexity indicators that could be applied in ANSP benchmarking. The complexity indicators are computed on a systematic basis for each day of the year. This annex presents for each ANSP the complexity score computed over the full year (365 days).

The complexity indicators are based on the concept of “interactions” arising when there are two aircraft in the same “place” at the same time. Hence, the complexity score is a measure of the potential number of interactions between aircraft defined as the total duration of all interactions (in minutes) per flight‐hour controlled in a given volume of airspace.

For each ANSP the complexity score is the product of two components:

The traffic density is expressed in adjusted density which measures the (uneven) distribution of traffic throughout the airspace (i.e. taking into account the relative concentration). The measure relies on dividing the airspace volume into a discrete grid of 20 nautical mile cells. For the purpose of this study, an interaction is defined as the simultaneous presence of two aircraft in a cell of 20x20 nautical miles and 3,000 feet in height.

The structural index originates from horizontal, vertical, and speed interactions and is computed as the sum of the three indicators.

Horizontal interactions indicator: A measure of the complexity of the flow structure based on the potential interactions between aircraft on different headings. The indicator is defined as the ratio of the duration of horizontal interactions to the total duration of all interactions.

Vertical interactions indicator: A measure of the complexity arising from aircraft in vertical evolution based on the potential interactions between climbing, cruising and descending aircraft. The indicator is defined as the ratio of the duration of vertical interactions to the total duration of all interactions

Speed interactions indicator: A measure of the complexity arising from the aircraft mix based on the potential interactions between aircraft of different speeds. The indicator is defined as the ratio of the duration of speed interactions to the total duration of all interactions

More information on the methodologies used for the computation of the complexity score in this report is available from the report on “Complexity Metrics for ANSP Benchmarking Analysis” available on the PRC webpage [Ref. 7].

Complexity score = Traffic density x Structural index

93

ANSP Complexity score (2014)

The complexity scores in the table below represent an annual average. Hence the complexity score in areas with a high level of seasonal variability may be higher during peak months.

* Note that ENAIRE’s (former Aena) complexity score is influenced by the low traffic density of Canarias

airspace.

Complexity Score 2014

ANSP and State vertical horizontal speed Total

A % change B C D E=B+C+D % change F= A * E % change

Skyguide (CH) 11.19 4.0% 0.26 0.61 0.22 1.10 -1.2% 12.30 2.7%

NATS (Continental) (UK) 10.09 1.4% 0.37 0.44 0.30 1.12 0.8% 11.28 2.2%

DFS (DE) 10.23 0.5% 0.26 0.57 0.24 1.08 -0.7% 11.01 ‐0.2%

Belgocontrol (BE) 7.55 5.0% 0.38 0.55 0.45 1.38 -1.3% 10.44 3.6%

MUAC (MUAC) 10.48 2.4% 0.26 0.55 0.17 0.98 0.8% 10.26 3.2%

LVNL (NL) 10.20 2.9% 0.18 0.43 0.39 1.00 0.0% 10.19 2.9%

ANS CR (CZ) 9.58 4.7% 0.13 0.53 0.16 0.83 -0.8% 7.92 3.8%

Austro Control (AT) 8.71 4.7% 0.17 0.53 0.19 0.88 -2.3% 7.68 2.3%

Slovenia Control (SI) 10.47 12.7% 0.09 0.54 0.10 0.73 -5.1% 7.63 7.0%

DSNA (FR) 10.51 4.5% 0.14 0.43 0.13 0.70 -0.7% 7.33 3.8%

DHMI (TR) 10.49 22.7% 0.14 0.31 0.20 0.65 -3.4% 6.84 18.5%

LPS (SK) 8.40 7.9% 0.08 0.48 0.14 0.71 -2.8% 5.94 4.8%

ENAV (IT) 5.72 7.1% 0.25 0.60 0.16 1.00 -1.4% 5.73 5.6%

HungaroControl (HU) 8.78 13.8% 0.05 0.46 0.13 0.64 -1.3% 5.64 12.4%

Croatia Control (HR) 8.25 3.4% 0.05 0.51 0.08 0.64 4.9% 5.26 8.4%

SMATSA (LY) 8.22 -3.5% 0.04 0.51 0.07 0.62 3.4% 5.11 ‐0.3%

ENAIRE (ES) 6.68 3.5% 0.14 0.37 0.13 0.64 0.0% 4.28 3.6%

BULATSA (BU) 8.63 24.7% 0.06 0.33 0.10 0.49 13.5% 4.26 41.5%

PANSA (PL) 4.72 -1.5% 0.14 0.53 0.22 0.89 0.3% 4.21 ‐1.1%

ROMATSA (RO) 7.47 27.0% 0.04 0.38 0.12 0.54 -9.7% 4.02 14.7%

SAKAERONAVIGATSIA (GE) 5.48 0.06 0.37 0.25 0.68 3.71

DCAC Cyprus (CY) 5.27 7.7% 0.16 0.39 0.11 0.66 4.9% 3.46 13.0%

NAVIAIR (DK) 3.56 0.4% 0.18 0.55 0.20 0.93 -1.4% 3.32 ‐1.0%

Albcontrol (AL) 6.65 0.7% 0.05 0.37 0.05 0.47 5.4% 3.13 6.2%

M‐NAV (MK) 5.45 30.9% 0.08 0.46 0.04 0.57 -3.4% 3.13 26.4%

LFV (SE) 3.04 1.1% 0.21 0.50 0.23 0.94 0.0% 2.87 1.1%

HCAA (GR) 4.55 8.0% 0.10 0.40 0.08 0.58 5.6% 2.66 14.0%

EANS (EE) 3.68 -2.6% 0.15 0.32 0.24 0.71 7.6% 2.60 4.8%

NAV Portugal (Continental) (PT) 4.23 12.3% 0.15 0.38 0.08 0.60 -0.4% 2.54 11.8%

LGS (LV) 3.25 -1.0% 0.09 0.48 0.16 0.72 2.4% 2.35 1.5%

Oro Navigacija (LT) 3.08 7.8% 0.08 0.47 0.16 0.71 -1.3% 2.19 6.4%

Avinor (Continental) (NO) 2.21 1.2% 0.27 0.46 0.26 0.99 -0.6% 2.18 0.6%

IAA (IE) 3.96 -2.4% 0.08 0.25 0.16 0.49 10.3% 1.95 7.6%

UkSATSE (UA) 2.72 -18.0% 0.06 0.38 0.17 0.61 -8.3% 1.64 ‐24.8%

Finavia (FI) 1.73 2.0% 0.25 0.33 0.35 0.93 -1.2% 1.61 0.8%

MATS (MT) 1.78 7.2% 0.06 0.36 0.23 0.65 3.2% 1.16 10.6%

MoldATSA (MD) 1.50 -38.4% 0.04 0.44 0.15 0.63 -7.8% 0.94 ‐43.2%

ARMATS (AM) 1.25 -5.9% 0.10 0.35 0.20 0.65 2.9% 0.81 ‐3.2%

Average 7.98 6.5% 0.19 0.46 0.18 0.82 ‐1.8% 6.55 4.6%

Adjusted

density

Structural index Complexity

Score

94

ANNEX IV - FRAMEWORK : SERVICE QUALITY

The conceptual framework for the analysis of ATM‐related service quality applied in this Performance Review Report (PRR) is illustrated below.

Conceptual framework for measuring ATM‐related service quality

The top level illustrates the passenger perspective which compares actual performance to published airline schedules (punctuality)40. Although on‐time performance is a valid measure from a passenger point of view, the involvement of many different stakeholders (airlines, ground handlers, airport operators, ATC, etc.) and the inclusion of time buffers in airline schedules to ensure a satisfactory on‐time performance require a more in depths analysis for the assessment of ANS‐related performance.

In order to better understand the ANS‐related contribution towards overall performance, the evaluation of ANS‐related service quality focuses on the “Efficiency” and the “Variability” of actual operations by phase of flight (see also grey information box). The flight phases where ANS is considered to have an impact on operations are highlighted in blue in the conceptual framework above.

Inefficiencies in the vertical flight profile for en‐route and in the TMA departure phase are presently not analysed in more detail in this report as they are generally not considered to be large contributors to ANS‐related inefficiencies. However, the magnitude can change by region or airport and it is acknowledged that there is scope for future improvement in those areas in order to get a more complete picture.

Although there are a number of factors outside the control of ANS (apron and stand limitations, etc.), a first evaluation of taxi‐in performance is provided in Chapter 5 for completeness reasons.

Efficiency and Variability

‘Efficiency’ in this report measures the difference between actual time/distance and an unimpeded reference time/distance. “Inefficiencies” can be expressed in terms of time and fuel and also have an environmental impact.

Due to inherent necessary (safety) or desired (noise, capacity, cost) limitations the reference values are not necessarily achievable at system level and therefore ANS‐related ‘inefficiencies” cannot be reduced to zero.

The “variability” of operations determines the level of predictability for airspace users and hence has an impact on airline scheduling and the ability to maximise the use of available resources (aircraft, crew, etc.). It focuses on the variance (distribution widths) associated with the individual phases of flight as experienced by airspace users. The higher the variability, the wider the distribution of actual travel times and the more time buffer is required in airline schedules to maintain a satisfactory level of punctuality.

40 It should be noted that there are differences between passenger‐centric and flight‐centric metrics. Flight centric

metrics do not reflect passenger disruption, such as missed connections and the average passenger delay is therefore typically higher than the average flight delay.

Departure delays

Airport Capacity• airport scheduling• achieved throughput• sustainability of ops.• etc.

En‐route efficiency

Origin airport

En‐route ATFM delays Airport

ATFM delays

Terminalefficiency

Reactionary delays

Management of arrival flows

Other (airline, airport, etc.)

Departure punctuality

Pre‐departuredelays(at gate)

Air Traffic Management

Weather

Taxi‐outefficiency

Scheduled block time (airlines) Buffer

Efficiency and variability of operations (ANS contribution)

Ground

Airborne

En‐route network Approach Arrival airport

Airline scheduling

ANS performance

Arrivalpunctuality

Passenger perspective

Taxi‐inefficiency

95

Not all delay is to be seen as negative. Due to the stochastic nature of air transport (weather conditions, etc.) and the way the system is operated today (airport slots, traffic flow management, etc.), different levels of “inefficiency” may be required or even desirable to maximise the use of scarce capacity at the given level of variability.

Imbalances between capacity and demand can occur en‐route and at airports due to a number of reasons including: weather, variability of traffic flows, scheduling, and the availability of ATC capacity and airspace.

A clear‐cut allocation between ANS (ATC capacity, staffing, etc.) and non‐ANS (weather, airline delay, etc.) related causes is often difficult. While ANS is often not the root cause of the problem, depending on the way those imbalances are managed and distributed along the various phases of flight (airborne vs. ground) through the application of flow measures, the outcome has a different impact on airspace users (punctuality, time, fuel burn, costs), the utilisation of capacity (en‐route and airport), and the environment (emissions).

For instance, whereas ATFM slots result in pre‐departure delays mainly experienced at the stands, inefficiencies in the gate‐to‐gate phase (taxi, en‐route, terminal) also generate additional fuel burn. Although keeping an aircraft at the gate saves fuel ‐ if it is held and capacity goes unused ‐ the cost to the airline of the extra delay may exceed the fuel cost by far.

The table below provides an overview of how ANS can impact on airspace users’ operations in terms of time, fuel burn and associated costs. The cost aspect of ANS‐related service quality is addressed in more detail in Chapter 2 and Annex V of this report.

ANS‐ related impact on airspace users’ operations

Impact on punctuality

Engine status

Impact on fuel burn/ CO2 emissions

Impact on airspace users’ costs

ANS related

inefficiencies At stand

Airport ATFM High OFF Quasi nil Time

En‐route ATFM

Gate‐to‐gate

Taxi phase Low/

moderate ON High Time + fuel En‐route phase

Terminal area

ANS‐related impact on airspace users’ operations

For ANS‐related delays at the gate (ATFM delays) the fuel burn is quasi‐nil but the level of predictability in the scheduling phase is low. Hence, the impact of ATFM delays on punctuality and associated costs to airspace users is significant (i.e. “tactical” delays) but the impact on fuel burn and the environment is negligible41.

ANS‐related inefficiencies in the gate‐to‐gate phase (taxi, en‐route, terminal holdings) are generally more predictable than ATFM delays as they are more related to inefficiencies embedded in the air route network or congestion levels which are similar every day. From an airspace user point of view, the impact on punctuality is usually limited as those inefficiencies are usually already embedded in the scheduled block times (“strategic delays”). However, the impact in terms of additional time, fuel, costs, and the environment is significant.

The high level analysis of service quality in Chapter 2 is supported by a more detailed analysis of operational en‐route ANS performance in Chapter 4 of this report. ANS‐related performance at airports is evaluated in more detail in Chapter 5.

41 It is acknowledged that in some cases aircraft operators try to make up for ATFM delay encountered at the

origin airport through increased speed which in turn may have a negative impact on total fuel burn for the entire flight.

96

ANNEX V - FRAMEWORK: ECONOMIC EVALUATION OF ANS PERFORMANCE

In Europe, airspace users bear the total economic costs of ANS services, which consist of ANS costs (en‐route and terminal) and quality of service related costs (due to ANS related inefficiencies). This Annex provides background information on the framework applied for the economic evaluation of ANS performance in Chapter 2 of this report.

The economic evaluation of ANS performance is an attempt to monetarise direct and indirect costs borne by airspace users in order to draw a consolidated high‐level picture.

As illustrated in the figure on the right side, while its primacy is fully recognised, it is not appropriate to include a monetary value for Safety. Hence the evaluation addresses ANS cost‐efficiency and ANS‐related operational performance, linked with demand capacity balancing.

Balancing capacity and demand

Insufficient capacity has a negative impact on ANS‐related service quality performance (high delays, etc.) and on airspace users’ costs; while the provision of capacity higher than demand contributes towards higher than necessary ANS charges (underutilisation of resources).

The total economic evaluation is useful to provide a consolidated high‐level view on ANS performance and to promote discussions on future ANS performance objectives as:

it allows comparability of the different metrics as all (but Safety) are expressed in monetary terms;

it is easy to understand at high level (e.g. policy makers, executives, media, etc);

it provides a high‐level view to assess the relative weight of the different KPAs and priorities for policy objectives; and,

it provides a high level framework to illustrate interdependencies and trade‐offs among KPAs.

However, the concept has also drawbacks which limit its suitability at local level and for target setting purposes:

it relies on assumptions for the monetarisation of the cost of delays and fuel incurred by airspace users;

trade‐offs will inevitably differ at a local/FAB level according to traffic characteristics, and the economic and working environment; and,

total economic costs do not indicate the scope for improvement in respective KPAs.

Chapter 2 of this report combines key results from the detailed analysis of ANS cost‐efficiency in Chapter 6 and the evaluation of operational performance en‐route and at airports in Chapter 4 and 5 respectively.

While the ANS en‐route and terminal costs can be easily taken from Chapter 6, estimating costs to airspace users as a result of ANS‐related inefficiencies is complex and requires expert judgement and assumptions, based on published statistics and robust data wherever possible. There are inevitably margins of uncertainty which need to be taken into account for the interpretation of the results.

Community perspective

Airspace User perspective

Air NavigationService providerPerspective

ANS charges

Service Quality (time, fuel inefficiency)

ANS Cost ‐efficiency

ECONOMY & ENVIRONMENT

Capacity

SAFETY

SafetyANS‐related operational performance

97

ANS‐related inefficiencies impact on airspace users in terms of cost of time and fuel.

The monetarisation of ANS‐related inefficiencies in terms of time in this report is based on the study from the University of Westminster [Ref. 11] which addresses estimated costs to airspace users. It does not address the wider costs of delay which may be applicable in contexts such as the full societal impact of delay.

Inefficiency costs are calculated separately for “strategic” delays (those accounted for in advance during the scheduling phase by adding buffer to the airline schedule) and “tactical” delays (those incurred on the day of operations and not accounted for in advance).

Costs of ANS‐related inefficiencies

The estimated airline delay costs in the University of Westminster study [Ref. 11] include direct costs (fuel, crew, maintenance, etc.) the network effect (i.e. cost of reactionary delays) and passenger related costs.

Whilst passenger ‘value of time’ is an important consideration in wider transport economics, only those costs which impact on the airline’s business (rebooking, compensation, market share and passenger loyalty related costs) were included in the estimate. Estimates of future emissions costs from the EU emission trading scheme from 01 January 2012 were not included.

Hence, in this report, en‐route and airport ATFM delays were considered as being “tactical” (infrequent with a low level of predictability) and inefficiencies in the gate‐to‐gate phase (taxi out, en‐route, terminal) were considered to be “strategic”.

Although the main share of ANS‐related inefficiencies in the gate‐to‐gate phase is largely predictable (route network, congestion, etc.), it is acknowledged that some of the inefficiencies are not fully predictable and therefore could be considered as being “tactical”. As there is presently no validated methodology for the quantification of “tactical” delay in the gate‐to‐gate phase, all inefficiencies in the gate‐to‐gate phase were considered to be “strategic” and therefore predictable during the scheduling phase.

COST OF “TACTICAL” DELAYS

“Tactical” delays occur infrequently and are therefore difficult to predict for airlines during the scheduling phase.

While the fuel burn is quasi nil, the impact on airspace users’ schedules is significant.

Due to the lower level of predictability and resulting passenger related (compensation, rebooking, etc.) and network (reactionary delay) related cost, time impact in terms of cost (one minute of “tactical” delay) is considered to be higher than for “strategic” delay (does not include costs for fuel burn).

The latest figures used for the computation of “tactical” delay in the report are provided in the adjacent grey box.

Cost of ATFM departure delays

Time: The “tactical” delay cost of one additional minute is estimated at €79 per minute (€2009 prices) on average for a flight in Europe (derived from [Ref. 11].

Due to the low level of predictability and resulting passenger and network costs, the cost of one additional minute of “tactical” delay is higher than the cost of one additional minute (strategic delay) embedded in the schedule (without fuel costs).

Fuel: Costs are negligible the delay is usually experienced at the gate with engines off.

98

COST OF “STRATEGIC” DELAYS

Although not entirely predictable, a large share of the time inefficiencies experienced every day in the gate‐to‐gate phase (taxi‐out, en‐route, terminal holdings) is already accounted for in the “strategic” phase and reflected in the scheduled block times which limits the impact on punctuality.

Due to the lower impact on airline schedule and therefore lower costs for compensation, rebooking, reactionary delay etc., the cost of time of one minute “strategic” delay is lower than for “tactical” delay.

Additionally to the cost of time, the significant cost of additional fuel burn needs to be considered.

Fuel price is a major driver of costs, especially in the context of increasing jet fuel prices. After the drop in 2009, average jet fuel price increased again between 2009 and 2012. Between 20013 and 2014 jet fuel price decreased again, reaching a level comparable to 2008 in 2014 (see grey box).

In view of the variation of jet fuel price over the past years and to enable time series analysis of ANS‐related performance, the computations for the economic evaluation of ANS performance in Chapter 2 were normalised.

In order to remove variations due to changes in jet fuel prices, the 2014 average jet fuel price was consistently used for all years. Hence, the “real” cost might have been higher or lower in the individual years, depending on how the 2014 price compares to the price in the respective year.

The latest figures used for the computation of “strategic” delay in the report are provided in the adjacent grey box.

Cost of ANS‐related inefficiencies in the gate to gate phase

The “strategic” delay costs in the gate‐to‐gate phase consist of a time and a fuel component.

Time: The “strategic” delay cost of one additional minute (without fuel) is estimated at €27 per minute (€2009 prices) on average for a flight in Europe (derived from [Ref. 11]).

Fuel: The fuel costs are based on the average annual spot price in 2012 expressed in (€2009 prices). The fuel price paid by airspace users was estimated to be 15% above the spot price and also includes a provision for fuel carriage penalties.

Based STATFOR statistics and the assumptions above, the average jet fuel price in 2013 was calculated at 701 € per tonne (€2009).

Jet fuel price

FURTHER CONSIDERATIONS

It is important to point out that there are inevitably margins of uncertainty in the approximation of delay costs. Although the consolidated view of ANS‐related costs to airspace provides a good high‐level estimate, it is acknowledged that there is scope for further refinements.

The information used for the analysis in Chapter 2 was derived from and should be read in conjunction with the analyses, assumptions and limitations detailed in Chapters 4 to 6 of this report.

0100200300400500600700800900

200

4

200

5

200

6

200

7

200

8

200

9

201

0

201

1

201

2

201

3

201

4

Avg

. pe

r to

n [€

2009

]

Estimated average jet fuel price paid by airspace users

Source: STATFOR

99

ANNEX VI - AVIATION RISK MODEL

The SESAR programme, the technological dimension of the SES initiative, develops and validates advanced ATM capabilities to ensure the sustainable growth of European aviation. SESAR also promotes global interoperability with the ICAO Aviation System Block Upgrades.

Given this dynamic environment, there is a need for States and Service Providers to proactively and predictively model safety risk. This ensures that safety is considered as an integrated part of the evolution of the Aviation System, and that safety levels will at least be maintained.

The ICAO Global Aviation Safety Plan, the European ATM Master Plan and the PRR 2013 both highlight the need for a proactive and predictive management of risk.

In Europe, these capabilities should provide an understanding of how ANS has contributed positively, as well as negatively, to aviation safety in the past, and therefore how the SESAR concepts could increase the former contribution and decrease the latter contribution.

Building upon longstanding research in aviation risk modelling carried out at the EUROCONTROL Research Centre, a model of aviation safety, the ATM “Integrated RISk ‐ the European Aviation Model” (IRIS) has been developed. IRIS shows how the various elements of the ATM System contribute to the overall achieved safety levels but also can project how a greatly modified ATM System could contribute in the future. IRIS provides the independent means for safety risk measurement in that it can be used:

to support the monitoring of Safety Performance Indicators (SPIs);

to enable these occurrences to be mitigated at European level;

to support the States in moving towards a performance‐based oversight;

to support the design and validation of ATM capabilities; and to monitor the safety progression while mitigation measures and advanced safety capabilities are progressively introduced.

For IRIS, several sources of data sources can be used for the provision of periodic SPIs and to facilitate the measurement of safety performance for ATM operations. The data sources can be clustered as follows:

The use, by ground and air industries, of automated safety data reporting/monitoring tools;

Occurrence reporting by Member States and air navigation service providers (voluntary or mandatory as per Regulation (EU) 376/2014);

Internal ANSP and operator (flight data monitoring) occurrence reporting;

Safety KPIs as per Regulation (EU) No 390/2013;

Data from conflict probing and resolution tools;

The analysis of tracking Data (from ANSP and airport operational units); and

Flight Planning Data (from Network Manager).

Based on the recently signed appendix to the Memorandum of Cooperation between the FAA and EUROCONTROL the two agencies will develop and use a shared risk management platform in the areas of: risk assessment; safety monitoring; safety improvement; and safe(r) evolution of the aviation system. The capabilities of the current EUROCONTROL ATM IRIS and the FAA “Integrated Safety Assessment Model” (ISAM), will provide an 'emergent risk monitoring' capability for ANSPs and FAA facilities. This shared capability will be developed and delivered in the next two years.

100

ANNEX VII - AIRPORT TRAFFIC AND AIRPORT ANS PERFORMANCE RECORDS

ICAO Name (IATA code)

Peak

declared

arrival

capacity

(Mvts/hr)

Peak

arrival

rate

(Mvts/hr)

Arrival

ATFM delay

(min/arr)

Add. ASMA

Time

(min/arr)

ATC

departure

delay

(min/dep)

Add. Taxi‐out

Time

(min/dep)

ATFM Slot

Adherence

Data source NM Apt. data CODA Apt. data NM

2014 vs. 2013 2014 vs. 2013 2014 2014 2014 2014 2014 2014 2014

EGLL London (LHR) 73.4 1.4% 472.8 0.2% 44 46 1.89 8.63 0.65 8.74 85%

LFPG Paris (CDG) 63.8 2.8% 471.3 ‐1.5% 62 64 0.28 0.77 0.55 3.92 84%

EDDF Frankfurt (FRA) 59.6 2.6% 469.0 ‐0.8% 55 60 1.30 2.61 1.19 3.63 91%

EHAM Amsterdam (AMS) 55.0 4.6% 448.8 3.0% 68 67 1.89 1.44 1.44 2.84 87%

LTBA Istanbul (IST) 56.8 10.6% 429.1 8.6% 29 38 1.72 N/A 3.34 N/A 74%

EDDM Munich (MUC) 39.7 2.7% 374.2 ‐1.3% 58 58 0.51 1.75 0.38 2.71 95%

LEMD Madrid (MAD) 41.8 5.3% 342.5 2.8% 48 44 0.16 0.91 0.81 3.92 96%

LIRF Rome (FCO) 38.5 6.5% 312.1 3.3% 54 49 0.32 1.72 2.18 6.79 90%

LEBL Barcelona (BCN) 37.5 6.6% 283.7 2.6% 38 36 0.47 1.89 0.44 3.92 95%

EGKK London (LGW) 38.1 7.5% 259.9 3.7% 30 29 0.69 N/A 0.72 N/A 89%

LSZH Zurich (ZRH) 25.4 2.5% 257.3 0.8% 36 38 2.71 3.16 2.24 2.86 90%

EKCH Copenhagen (CPH) 25.6 6.5% 251.7 2.8% 52 35 0.04 1.35 0.05 1.74 93%

ENGM Oslo (OSL) 24.1 4.8% 247.7 2.7% 34 34 0.79 N/A 0.07 N/A 98%

LOWW Vienna (VIE) 22.5 2.2% 247.2 ‐0.2% 48 42 0.83 1.98 0.45 2.16 86%

LFPO Paris (ORY) 28.9 2.1% 231.1 ‐1.1% 34 33 0.87 1.01 0.37 2.00 82%

ESSA Stockholm (ARN) 22.5 8.5% 228.3 3.8% 42 35 0.27 1.08 0.07 3.13 97%

EBBR Brussels (BRU) 21.9 14.7% 225.4 6.7% 48 41 0.88 0.88 0.56 2.39 95%

EDDL Dusseldorf (DUS) 21.9 2.9% 210.1 ‐0.1% 33 33 0.39 1.82 0.52 2.62 93%

LSGG Geneva (GVA) 15.1 5.1% 181.1 2.0% 22 24 1.50 2.04 0.30 2.88 92%

EDDT Berlin (TXL) 20.7 5.6% 180.3 4.3% 30 29 0.44 N/A 0.56 N/A 91%

LTFJ Istanbul (SAW) 23.6 25.4% 179.6 23.9% N/A 23 2.78 N/A 1.62 N/A 63%

EIDW Dublin (DUB) 21.7 7.7% 179.2 5.8% 29 22 0.05 1.50 0.54 3.30 96%

LTAI Antalya (AYT) 28.5 4.1% 172.6 4.1% N/A 30 0.26 N/A 1.07 N/A 67%

LEPA Palma (PMI) 23.1 1.5% 172.4 1.8% 33 34 0.93 1.34 0.45 2.73 97%

EGCC Manchester (MAN) 22.1 5.8% 170.4 0.9% 30 25 0.07 1.56 0.56 N/A 86%

EFHK Helsinki (HEL) 15.9 4.4% 168.0 ‐0.1% 48 39 0.19 0.95 0.18 2.04 90%

LIMC Milan (MXP) 18.8 5.0% 166.6 1.0% 40 24 0.01 0.58 0.56 3.37 97%

LPPT Lisbon (LIS) 18.1 13.3% 156.6 7.3% 26 22 0.63 1.31 0.92 2.67 84%

EGSS London (STN) 19.9 11.8% 155.9 8.9% 28 21 0.08 0.83 0.61 2.23 90%

LGAV Athens (ATH) 15.2 21.3% 148.8 9.9% 35 21 0.53 0.32 1.18 90%

EDDH Hamburg (HAM) 14.8 9.3% 146.6 7.2% 27 24 0.28 1.27 0.22 1.97 91%

EPWA Warsaw (WAW) 10.6 ‐0.9% 138.5 ‐2.5% 28 22 0.32 N/A 0.27 2.94 90%

LFMN Nice (NCE) 11.7 0.9% 136.7 ‐2.6% 30 25 0.32 N/A 0.50 N/A 82%

LKPR Prague (PRG) 11.1 1.6% 121.2 ‐2.9% 33 24 0.19 1.23 0.57 1.86 93%

EDDK Cologne (CGN) 9.5 4.1% 120.4 2.7% 40 18 0.02 0.63 0.34 1.32 94%

EDDS Stuttgart (STR) 9.7 1.5% 114.0 ‐0.2% 32 20 0.08 0.72 0.14 N/A 97%

LIML Milan (LIN) 9.0 0.0% 111.7 ‐0.1% 20 18 0.01 0.47 0.44 2.15 97%

LFLL Lyon (LYS) 8.5 ‐1.1% 108.1 ‐6.9% 36 27 0.06 0.69 0.15 0.97 90%

EGPH Edinburgh (EDI) 10.2 3.9% 107.8 ‐2.1% N/A 17 0.01 0.72 0.19 1.77 90%

LEMG Malaga (AGP) 13.7 6.4% 105.6 5.1% 28 21 0.02 0.88 0.40 2.33 89%

EGGW London (LTN) 103.5 6.6% 22 16 0.05 N/A 0.70 3.38 82%

GCLP Las Palmas (LPA) 10.3 5.6% 99.6 6.1% 24 20 0.03 1.22 0.40 1.68 87%

ENBR Bergen (BGO) 6.2 ‐0.9% 97.6 ‐2.3% 15 17 0.56 N/A 0.00 N/A 96%

LFML Marseille (MRS) 8.2 ‐0.9% 97.2 ‐5.6% N/A 15 0.20 N/A 0.16 N/A 81%

EGBB Birmingham (BHX) 9.7 6.4% 95.6 4.2% N/A 15 0.03 0.60 0.17 1.62 81%

LROP Bucharest (OTP) 8.3 8.8% 91.2 4.2% N/A 14 N/A 0.19 N/A 93%

LTAC Esenboğa (ESB) 11.0 0.9% 89.3 ‐2.3% N/A 16 0.05 N/A 0.19 N/A 60%

LFBO Toulouse (TLS) 7.5 ‐0.7% 88.7 ‐3.1% N/A 15 0.09 N/A 0.23 N/A 92%

LHBP Budapest (BUD) 9.1 7.5% 86.4 3.4% 26 15 0.63 0.17 1.00 95%

ENZV Stavanger (SVG) 4.7 0.3% 82.1 4.0% 15 16 0.31 N/A 0.01 N/A 96%

EGPF Glasgow (GLA) 7.7 4.8% 79.6 2.3% 0 14 N/A 0.17 N/A 92%

LIPZ Venice (VCE) 8.5 0.9% 77.7 ‐3.9% 16 15 0.11 N/A 0.66 1.28 90%

LFSB Basel‐Mulhouse (BSL) 6.5 11.1% 75.9 4.3% N/A 15 0.23 N/A 0.13 1.81 87%

EGLC London (LCY) 75.7 2.8% 19 18 1.35 N/A 0.62 2.22 89%

LTBJ Izmir (ADB) 11.0 6.8% 75.1 5.3% N/A 13 N/A 0.53 N/A 64%

EGPD Aberdeen (ABZ) 3.8 8.0% 73.0 3.3% N/A 14 0.27 N/A 0.07 3.06 88%

UKBB Kiev (KBP) 6.9 ‐13.1% 72.6 ‐10.9% 48 21 0.04 N/A 0.14 N/A 81%

LIME Bergamo (BGY) 8.8 ‐2.1% 67.5 ‐5.2% 16 13 N/A 0.75 2.21 94%

EVRA Riga (RIX) 4.8 0.4% 65.4 ‐2.7% N/A 21 N/A 0.03 1.95 94%

LYBE Beograd (BEG) 4.6 30.9% 62.2 24.3% N/A 14 0.02 N/A 0.13 N/A 86%

ELLX Luxembourg (LUX) 2.5 12.3% 59.1 2.6% 15 14 0.08 N/A 0.15 N/A 83%

LCLK Larnaca (LCA) 5.3 7.6% 43.6 7.5% 12 12 0.01 N/A 0.34 N/A 82%

LBSF Sofia (SOF) 3.8 8.9% 41.2 3.7% 11 8 N/A 0.19 1.16 98%

LMML Malta (MLA) 4.4 7.3% 38.4 8.0% N/A 9 0.02 N/A 0.14 N/A 96%

EYVI Vilnius (VNO) 3.0 10.7% 36.5 13.9% N/A 7 N/A 0.03 N/A 90%

LDZA Zagreb (ZAG) 2.4 5.7% 36.1 3.8% N/A 8 0.01 N/A 0.13 N/A 89%

EETN Tallinn (TLL) 2.0 3.2% 34.4 0.0% N/A 9 N/A 0.03 N/A 94%

LJLJ Ljubljana (LJU) 1.3 3.1% 26.1 ‐4.3% 20 8 N/A 0.08 N/A 94%

UDYZ Yerevan (EVN) 20.9 19.4% N/A 5 N/A 0.20 N/A 77%

UGTB Tbilisi (TSB) 20.1 N/A 6 N/A 0.81 N/A N/A

LUKK Kishinev (KIV) 1.8 34.8% 19.3 12.4% N/A 6 N/A 0.13 N/A 96%

LATI Tirana (TIA) 1.8 3.0% 18.1 ‐9.4% N/A 6 0.04 N/A 0.13 N/A 81%

LZIB Bratislava (BTS) 0.0 17.9 ‐2.6% N/A 6 N/A 0.13 N/A 94%

LWSK Skopje (SKP) 1.2 23.0% 12.9 16.5% 20 4 N/A 0.04 N/A 91%

LQSA Sarajevo (SJJ) 0.7 6.6% 11.3 8.3% N/A 5 N/A 0.03 N/A 75%

LYPG Podgorica (TGD) 0.7 1.6% 10.1 ‐6.2% N/A 4 N/A 0.33 N/A 97%

DEPARTURE TRAFFIC FLOW

Passengers ('M)

ACI

IFR movements

('000)

NM

ARRIVAL TRAFFIC FLOW

101

ANNEX VIII - RECONCILIATION WITH PRB 2013 MONITORING REPORT

Actual 2013 data

The actual 2013 data used in the PRR report are based on States November 2014 submission to the enlarged Committee for Route Charges.

The table below shows how this information reconciles with what was reported in the context of the PRB 2013 Monitoring Report (Figure 25, p. 40):

RP1 SES States ‐ reconciliation with PRB annual Monitoring Report

The actual 2013 en‐route costs for the RP1 SES States are lower by ‐30.9 M€2009 in the PRR report (November 2014 data) compared to the figure published in the PRB 2013 Monitoring Report (June 2014 data) due to the downwards revision of actual costs by DSNA (France) by ‐30.7 M€2009 and by the German NSA by ‐0.1 M€2009.

RP2 (2015‐2019)

The data for SES States reflects latest available data on en‐route costs provided up to 15 January 2015. The table below shows how this information reconciles with what was reported in the context of the PRB RP2 Assessment Report (Table 20, p. 37):

RP1 SES States – reconciliation with PRB annual Monitoring Report

The following States have provided revised cost data for RP2 after the submission of their initial FAB RP2 Performance Plans: Austria, Bulgaria, Croatia, Czech Republic, Denmark, Finland, Germany, Hungary, Ireland, Malta, Netherlands, Norway, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden and UK.

En route total costs (RP1) - MEUR2009 2013APRB Annual Monitoring Report 2013 5.978,8 PRR 2014 5.947,9 Difference 30,9-

En route total costs (RP2) - MEUR2009 2014 F 2015 D 2016 D 2017 D 2018 D 2019 DPRB RP2 Assessment Report (6/10/2014) 6.242,3 6.271,8 6.257,8 6.249,9 6.206,0 6.158,8 PRR 2014 6.227,3 6.239,6 6.183,8 6.180,4 6.139,5 6.092,3 Difference 14,9- 32,2- 74,0- 69,4- 66,5- 66,5-

102

ANNEX IX - GLOSSARY

ACC Area Control Centre. That part of ATC that is concerned with en‐route traffic coming from or going to adjacent centres or APP. It is a unit established to provide air traffic control service to controlled flights in control areas under its jurisdiction.

Accident (ICAO Annex 13)

An occurrence associated with the operation of an aircraft which takes place between the time any person boards the aircraft with the intention of flight until such time as all such persons have disembarked, in which: a) a person is fatally or seriously injured as a result of:

Being in the aircraft, or

Direct contact with any part of the aircraft, including parts which have become detached from the aircraft, or

Direct exposure to jet blast, except when the injuries are from natural causes, self‐inflicted or inflicted by other persons, or when the

injuries are to stowaways hiding outside the areas normally available to the passengers and crew; or b) the aircraft sustains damage or structural failure which:

Adversely affects the structural strength, performance or flight characteristics of the aircraft, and

Would normally require major repair or replacement of the affected component, except for engine failure or damage, when the damage is limited to the engine, its cowlings or accessories, or for damage limited to propellers, wing tips, antennas, tyres, brakes, fairings, small dents or puncture holes in the aircraft skin;

c) the aircraft is missing or completely inaccessible.

A‐CDM Airport Collaborative Decision‐Making

ACE Reports Air Traffic Management Cost‐Effectiveness (ACE) Benchmarking Reports

ACI Airports Council International (http://www.aci‐europe.org/)

AEA Association of European Airlines (http://www.aea.be)

Aena see ENAIRE

Agency The EUROCONTROL Agency

AIP Aeronautical Information Publication

Airside The aircraft movement area (stands, apron, taxiway system, runways etc.) to which access is controlled.

Airspace Infringement

(also known as unauthorised penetration of airspace). The penetration by an aircraft into a portion of airspace without prior permission of the appropriate authorities (when such prior permission is required). EUROCONTROL HEIDI – ESARR 2 taxonomy

AIS Aeronautical Information Service

ALAQS EUROCONTROL Airport Local Air Quality Studies

AMAN Arrival Management Function

ANS Air Navigation Service. A generic term describing the totality of services provided in order to ensure the safety, regularity and efficiency of air navigation and the appropriate functioning of the air navigation system.

ANS CR Air Navigation Services of the Czech Republic. ANS Provider ‐ Czech Republic.

ANSP Air Navigation Services Provider

AO Aircraft Operator

APP Approach Control Unit

APU Auxiliary Power Units

ARMATS Armenian Air Traffic Services, ANS Provider ‐ Armenia

ARN V8 ATS Route Network (ARN) ‐ Version 8

ASK Available seat‐kilometres (ASK): Total number of seats available for the transportation of paying passengers multiplied by the number of kilometres flown

ASM Airspace Management

ASMA Arrival Sequencing and Metering Area

ASMT EUROCONTROL Automatic Safety Monitoring Tool

AST Annual Summary Template

ATC Air Traffic Control. A service operated by the appropriate authority to promote the safe, orderly and expeditious flow of air traffic.

ATCO Air Traffic Control Officer

ATFCM Air Traffic Flow and Capacity Management.

103

ATFM Air Traffic Flow Management. ATFM is established to support ATC in ensuring an optimum flow of traffic to, from, through or within defined areas during times when demand exceeds, or is expected to exceed, the available capacity of the ATC system, including relevant aerodromes.

ATFM delay (NMD definition)

The duration between the last Take‐Off time requested by the aircraft operator and the Take‐Off slot given by the EUROCONTROL Network Management Directorate

ATFM Regulation When traffic demand is anticipated to exceed the declared capacity in en‐route control centres or at the departure/arrival airport, ATC units may call for “ATFM regulations”.

ATK Available tonne kilometres (ATK) is a unit to measure the capacity of an airline. One ATK is equivalent to the capacity to transport one tonne of freight over one kilometre.

ATM Air Traffic Management. A system consisting of a ground part and an air part, both of which are needed to ensure the safe and efficient movement of aircraft during all phases of operation. The airborne part of ATM consists of the functional capability which interacts with the ground part to attain the general objectives of ATM. The ground part of ATM comprises the functions of Air Traffic Services (ATS), Airspace Management (ASM) and Air Traffic Flow Management (ATFM). Air traffic services are the primary components of ATM.

ATMAP ATM Performance at Airports

ATS Air Traffic Service. A generic term meaning variously, flight information service, alerting service, air traffic advisory service, air traffic control service.

Austro Control Austro Control: Österreichische Gesellschaft für Zivilluftfahrt mbH, ANS Provider ‐ Austria

AVINOR ANS Provider ‐ Norway

Belgocontrol ANS Provider ‐ Belgium

BULATSA Air Traffic Services Authority of Bulgaria. ANS Provider ‐ Bulgaria.

CAA Civil Aviation Authority

CANSO Civil Air Navigation Services Organisation (http://www.canso.org)

CDA Continuous Descent Approach

CDM Collaborative Decision Making

CDR Conditional Routes

CEF Capacity Enhancement Function

CFMU (See NMD) Formerly the EUROCONTROL Central Flow Management Unit. Now the EUROCONTROL Network Management Directorate (NMD)

CLR Deviation from ATC clearance

CMA Continuous Monitoring Approach (ICAO USOAP Cycle)

CNS Communications, Navigation, Surveillance.

CO2 Carbon dioxide

CODA EUROCONTROL Central Office for Delay Analysis

Composite flight hour En‐route flight hours plus IFR airport movements weighted by a factor that reflected the relative importance of terminal and en‐route costs in the cost base (see ACE reports)

CRCO EUROCONTROL Central Route Charges Office

Croatia Control Hrvatska kontrola zračne plovidbe d.o.o. ANS Provider ‐ Croatia,

CTOT Calculated Take‐Off Time

Dangerous Phenomena

The principal dangerous weather phenomena are: Cumulonimbus (CB) with or without precipitation, Tower Cumulus (TCU), Thunder with or without precipitation (TS) , Ice Pellets (PL),Small Hail and/or Snow Pellets (GS); Hail (GR), Funnel cloud (tornado or waterspout) (FC) , Squall (SQ) , Volcanic Ash (VA), Dust‐storm (DS), Sandstorm (SS), Sand (SA), Dust/sand whirls (PO)

DCAC Cyprus Department of Civil Aviation of Cyprus. ANS Provider ‐ Cyprus.

DFS DFS Deutsche Flugsicherung GmbH, ANS Provider ‐ Germany

DGCA Directors General of Civil Aviation

DHMi Devlet Hava Meydanlari Isletmesi Genel Müdürlügü (DHMi), General Directorate of State Airports Authority, Turkey. ANS Provider – Turkey.

DSNA Direction des Services de la Navigation Aérienne. ANS Provider ‐ France

DSS/OVS/SAF Unit EUROCONTROL Directorate Single Sky/Oversight/Safety Unit. Formerly the Safety Regulation Unit.

DUR Determined Unit Rate

EAD European AIS Database

EANS Estonian Air Navigation Services. ANS Provider – Estonia.

EAPPRI European Action Plan for the Prevention of Runway Incursions

EASA European Aviation Safety Agency

EATM European Air Traffic Management (EUROCONTROL)

EATMN European Air Traffic Management Network (SES legislation)

EC European Commission

ECAC European Civil Aviation Conference.

ECCAIRS European accident and incident database

104

ECTL Acronym for EUROCONTROL

EEA European Economic Area (EU Member States + Iceland, Norway and Lichtenstein)

EEA European Environmental Agency

Effective capacity The traffic level that can be handled with optimum delay (cf. PRR 5 (2001) Annex 6)

ENAIRE Formerly AENA ‐ ANS Provider ‐ Spain

ENAV ENAV S.p.A. ‐ Italian Company for Air Navigation Services

EoSM Effectiveness of Safety Management

ERA European Regional Airlines Association (http://www.eraa.org)

ESRA 2008 Area European Statistical Reference Area (see STATFOR Reports) Albania, Austria, Belgium, Bosnia and Herzegovina, Bulgaria, Canary Islands, Croatia, Cyprus, Czech Republic, Denmark, Finland, France, FYROM, Germany, Greece, Hungary, Ireland, Italy, Lisbon FIR, Luxembourg, Malta, Moldova, Montenegro, Netherlands, Norway, Poland, Romania, Santa Maria FIR, Serbia, Slovak Republic , Slovenia, Spain, Sweden, Switzerland, Turkey, Ukraine, United Kingdom

ESSIP European Single Sky ImPlementation plan

EU European Union

EU States The member States of the European Union.

EU‐ETS Emissions Trading Scheme. The objective of the EU ETS is to reduce greenhouse gas emissions in a cost‐effective way and contribute to meeting the EU’s Kyoto Protocol targets.

EUROCONTROL The European Organisation for the Safety of Air Navigation. It comprises Member States and the Agency.

EUROCONTROL Member States (40 States in 2014)

Albania, Armenia, Austria, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Finland, France, Georgia, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Moldova, Monaco, Montenegro, Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, The former Yugoslav Republic of Macedonia, Turkey, Ukraine and United Kingdom of Great Britain and Northern Ireland

EUROCONTROL Route Charges System

A regional cost‐recovery system that funds air navigation facilities and services and supports Air Traffic Management developments. It is operated by the EUROCONTROL Central Route Charges Office (CRCO), based in Brussels. www.eurocontrol.int/crco

EUROSTAT The Statistical Office of the European Community

FAB Functional Airspace Blocks

FABEC States Belgium, France, Germany, Luxembourg, the Netherlands and Switzerland

FINAVIA ANS provider – Finland

FIR Flight Information Region. An airspace of defined dimensions within which flight information service and alerting service are provided.

FL Flight Level. Altitude above sea level in 100 feet units measured according to a standard atmosphere. Strictly speaking a flight level is an indication of pressure, not of altitude. Only above the transition level (which depends on the local QNH but is typically 4000 feet above sea level) flight levels are used to indicate altitude, below the transition level feet are used.

FMP Flow Management Position

FUA Flexible Use of Airspace

FYROM Former Yugoslav Republic of Macedonia

GA (General Aviation)

All civil aviation operations other than scheduled air services and non‐scheduled air transport operations for remuneration or hire.

GAT General Air Traffic. Encompasses all flights conducted in accordance with the rules and procedures of ICAO. PRR 2014 uses the same classification of GAT IFR traffic as STATFOR:

GCD Great Circle Distance

GDP Gross Domestic Product

HCAA Hellenic Civil Aviation Authority. ANS Provider ‐ Greece

HungaroControl ANS Provider ‐ Hungary

IAA Irish Aviation Authority. ANS Provider ‐ Ireland

IATA International Air Transport Association (www.iata.org)

ICAO International Civil Aviation Organization

IFR Instrument Flight Rules. Properly equipped aircraft are allowed to fly under bad‐weather conditions following instrument flight rules.

Incident (ICAO Annex 13)

An occurrence, other than an accident, associated with the operation of an aircraft which affects or could affect the safety of operation.

Incident Category A (ICAO Doc 4444)

A serious incident: AIRPROX ‐ Risk Of Collision: “The risk classification of an aircraft proximity in which serious risk of collision has existed”.

Incident Category B (ICAO Doc 4444)

A major incident. AIRPROX ‐ Safety Not Assured: “The risk classification of an aircraft proximity in which the safety of the aircraft may have been compromised”.

IS Inadequate separation

JC The EUROCONTROL definition of “just culture”, also adopted by other European aviation

105

Just culture stakeholders, is a culture in which “front line operators or others are not punished for actions, omissions or decisions taken by them that are commensurate with their experience and training, but where gross negligence, wilful violations and destructive acts are not tolerated.”

JRC EC Joint Research Centre

KPA Key Performance Area

KPI Key Performance Indicator

LAQ Local Air Quality

LFV Luftfartsverket. ANS Provider ‐ Sweden.

LGS SJSC Latvijas Gaisa Satiksme (LGS). ANS Provider ‐ Latvia

LPS Letové Prevádzkové Služby. ANS Provider ‐ Slovak Republic

LSSIP Local Single Sky ImPlementation plans/reports (formerly Local Convergence and Implementation Plans)

LTO Landing and Take‐off Cycle

LVNL Luchtverkeersleiding Nederland. ANS Provider ‐ Netherlands

Maastricht UAC The EUROCONTROL Upper Area Centre (UAC) Maastricht. It provides ATS in the upper airspace of Belgium, Luxembourg, Netherlands and Northern Germany.

MAC Mid‐air collision

MATS Malta Air Traffic Services Ltd. ANS Provider ‐ Malta

MET Meteorological Services for Air Navigation

METAR Meteorological Terminal Aviation Routine Weather Report or Meteorological Aerodrome Report

MIL Military flights

M‐NAV M‐NAV ‐ Macedonian Air Navigation Service Provider, PCL. ANS provider in the Republic of Macedonia

MoldATSA Moldavian Air Traffic Services Authority. ANS Provider ‐ Moldova

MTOW Maximum Take‐off Weight

MUAC Maastricht Upper Area Control Centre, EUROCONTROL

NATA Albania National Air Traffic Agency. ANS Provider ‐ Albania

NATS National Air Traffic Services. ANS Provider ‐ United Kingdom

NAV Portugal Navegação Aérea de Portugal – NAV Portugal, E.P.E.

NAVIAIR Naviair, Air Navigation Services. ANS Provider – Denmark

NERL NATS (En Route) Limited

NM Nautical mile (1.852 km)

NM Network Manager

NMD EUROCONTROL Network Management Directorate (formerly the EUROCONTROL Central Flow Management Unit ‐ CFMU).

NO2 Nitrogen dioxide

NOP Network Operations Plan

NSA National supervisory Authorities

OAT Operational Air Traffic

Occurrence (Source: ESARR 2)

Accidents, serious incidents and incidents as well as other defects or malfunctioning of an aircraft, its equipment and any element of the Air Navigation System which is used or intended to be used for the purpose or in connection with the operation of an aircraft or with the provision of an air traffic management service or navigational aid to an aircraft.

OPS Operational Services

Organisation See “EUROCONTROL”.

Oro Navigacija State Enterprise Oro Navigacija. ANS Provider ‐ Lithuania

PANSA Polish Air Navigation Services Agency. ANS Provider ‐ Poland

Passenger Load factor

Revenue passenger‐kilometres (RPK) divided by the number of available seat‐kilometres (ASK).

PC Provisional Council of EUROCONTROL

Permanent Commission

The governing body of EUROCONTROL. It is responsible for formulating the Organisation’s general policy.

PI Performance Indicator

PRB Performance Review Body of the Single European Sky

PRC Performance Review Commission

Primary Delay A delay other than reactionary

Productivity Hourly productivity is measured as Flight‐hours per ATCO‐hour (see ACE reports)

PRR Performance Review Report (i.e. PRR 2014 covering the calendar year 2014)

PRU Performance Review Unit

R&D Research & Development

RAD Route availability document

RAT Risk Analysis Tool for Safety

106

Reactionary delay Delay caused by late arrival of aircraft, crew, passengers or baggage from previous journeys

Revised Convention Revised EUROCONTROL International Convention relating to co‐operation for the Safety of Air Navigation of 13 December 1960, as amended, which was opened for signature on 27 June 1997.

RI Runway incursion: Any unauthorised presence on a runway of aircraft, vehicle, person or object where an avoiding action was required to prevent a collision with an aircraft. Source: ESARR 2.

ROMATSA Romanian Air Traffic Services Administration. ANS Provider ‐ Romania

RP1 First Reference Period (2012‐2014) of the SES Performance Scheme

RP2 Second Reference Period (2015‐2019) of the SES Performance Scheme

RPK Revenue passenger‐kilometre (RPK): One fare‐paying passenger transported one kilometre.

RTK Revenue Tonne Kilometre

RVSM Reduced Vertical Separation Minima

Serious incident (ICAO Annex 13)

An incident involving circumstances indicating that an accident nearly occurred.

SES Single European Sky (EU)

SES States The 27 EU States (see “EU States” above) plus Norway and Switzerland

SESAR The Single European Sky ATM Research programme

Severity The severity of an accident is expressed according to:

the level of damage to the aircraft (ICAO Annex 13 identifies four levels: destroyed: substantially destroyed, slightly damaged and no damage);

the type and number of injuries (ICAO Annex 13 identifies three levels of injuries: fatal, serious and minor/none).

PRRs focus on Severity A (Serious Incident) and Severity B (Major Incident).

SFMS Framework Maturity Survey (SFMS

Skyguide ANS Provider ‐ Switzerland

Slot (ATFM) A take‐off time window assigned to an IFR flight for ATFM purposes

Slovenia Control ANS Provider ‐ Slovenia

SM Separation Minima is the minimum required distance between aircraft. Vertically usually 1000 feet below flight level 290, 2000 feet above flight level 290. Horizontally, depending on the radar, 3 NM or more. In the absence of radar, horizontal separation is achieved through time‐separation (e.g. 15 minutes between passing a certain navigation point).

SMATSA Serbia and Montenegro Air Traffic Services Agency

SMI Separation Minima Infringement: A situation in which prescribed separation minima were not maintained between aircraft.

SMI Separation minima infringement.

SMS Safety Management System

SOx Sulphur oxide gases

SPI Safety Performance Indicator

SRC Safety Regulation Commission

SRU (see DSS/OVS/SAF)

SSC Single Sky Committee

SSP State Safety Programme

STATFOR EUROCONTROL Statistics & Forecasts Service

SU Service Units

SUA Special Use Airspace

TCZ Terminal Charging Zone

TMA Terminal manoeuvring area

TOBT Target Off‐Block Time

TRA Temporary Reserved Area

TSA Temporary Segregated Area

TSAT Target Start‐up Approval Time

UAC Upper Airspace Area Control Centre

UAP Unauthorised penetration of airspace (also known as Airspace Infringement). The penetration by an aircraft into a portion of airspace without prior permission of the appropriate authorities (when such prior permission is required). EUROCONTROL HEIDI – ESARR 2 taxonomy

UK CAA United Kingdom Civil Aviation Authority

UK NATS United Kingdom National Air Traffic Services

UkSATSE Ukrainian State Air Traffic Service Enterprise. ANS Provider ‐ Ukraine

UR Unit Rate

USOAP ICAO Universal Safety Oversight Audit Programme

UUP Updated Airspace Use Plan

VFR Visual Flight Rules

107

ANNEX X - REFERENCES

PRC documentation can be consulted and downloaded from the PRC website

http://www.EUROCONTROL.int/prc

1 Article 1 of the PRC’s Terms of Reference, adopted in 2003.

2 Performance Review Commission, Performance Review Report 2013 (PRR 2013). An assessment of Air Traffic Management in Europe during the calendar year 2013 (May 2014).

3 Commission Decision of 29.7.2010 on the designation of the Performance Review Body of the Single European Sky C (2010)5134 final.

4 EUROCONTROL STATFOR 7‐year forecast, February 2014, https://www.eurocontrol.int/articles/forecasts.

5 EUROCONTROL STATFOR 7‐year forecast, March 2015, https://www.eurocontrol.int/articles/forecasts.

6 Association of European Airlines (2015), http://www.aea.be/component/newsletter/newsletter/85.html

7 Performance Review Commission. “Complexity Metrics for ANSP Benchmarking Analysis”. Report by the ACE Working Group on complexity, 2006.

8 http://www.flightglobal.com/news/articles/2014‐shows‐lowest‐airline‐fatal‐accident‐rate‐in‐history‐407559/.

9 Performance Review Commission, Performance Review Report 2012 (PRR 2012). An assessment of Air Traffic Management in Europe during the calendar year 2012 (May 2013).

10 European Environment Agency (EEA), Greenhouse Gas Emissions by IPCC sector available at:

http://dataservice.eea.europa.eu/pivotapp/pivot.aspx?pivotid=475

11 European airline delay cost reference values (University of Westminster), Final report (Version 3.2), (March, 2011).

12 Safety Review Commission, "Annual Safety Report for 2013", (February, 2014).

13 European Single Sky Implementation Plan (ESSIP) 2014, the LSSIP reports, and information available in the LSSIP database (2014); https://www.eurocontrol.int/articles/lssip.

14 Performance Review Commission “Evaluation of Vertical flight‐efficiency” (March 2008).

15 http://www.eurocontrol.int/prudata/pru_meta.html;

16 Grewe, V., Champougny, T., Matthes, S., Frömming, C., Brinkop, S., Søvde, A.O., Irvine,E.A., Halscheidt, L., Reduction of the air traffic's contribution to climate change: A REACT4C case study, 10.1016/j.atmosenv.2014.05.059, Atmos. Environm. 94, 616‐625, 2014c.

17 European Route Network Improvement Plan (ERNIP) 2013‐2015.

18 Commission Regulation (EC) No 2150/2005 of 23 December 2005 laying down common rules for the flexible use of airspace OJ, 24.12.2005.

19 Commission Implementing Regulation (EU) No 677/2011 of 7 July 2011 laying down the detailed rules for the implementation of air traffic management (ATM) network functions and amending Regulation (EU) No 691/2010. OJ L 96, 31.3.2004, p.1.

20 Flight Plan & ATFCM Adherence campaign initiated by DMEAN and the Network Operations Centre (CFMU) in April 2009.

21 EUROCONTROL, “Challenges of growth 2013” report.

22 Performance Review Commission and FAA‐ATO, “U.S./Europe Comparison of ATM‐Related Operational Performance 2013”, (June 2014).

23 Commission Regulation (EC) No 255/2010 of 25 March 2010 laying down common rules on air traffic flow management. OJ L 80, 26.3.2010, p.10.

108

24 PRB RP1 Annual Monitoring Report 2013 (6 October 2014), http://www.eurocontrol.int/prb/publications.

25 PRB Assessment of FAB Performance Plans for RP2 (October 2014), http://www.eurocontrol.int/prb/publications.

26 Commission Implementing Regulation (EU) No 391/2013 of 3 May 2013 laying down a common charging scheme for air navigation services. OJ L 128, 9.5.2013, p31.

27 Commission Implementing Regulation (EU) No 390/2013 of 3 May 2013 laying down a performance scheme for air navigation services and network functions, OJ L 128, 9.5.2013, p. 1.

28 Commission Regulation (EC) No 1794/2006 of 6 December 2006 laying down a common charging scheme for air navigation services. OJ L 341, 7.12.2006, p. 3‐16.

29 Directive 2009/12/EC of the European Parliament and of the Council of 11 March 2009 on airport charges, OJ L 70, 14.3.2009, p. 11.

30 ATM Cost‐effectiveness (ACE) 2013 Benchmarking Report. Report commissioned by the Performance Review Commission (May 2015).

About the Performance Review Commission

The Performance Review Commission (PRC) provides independent advice on European Air Traffic Management (ATM) Performance to the EUROCONTROL Commission through the Provisional Council.

The PRC was established in 1998, following the adoption of the European Civil Aviation Conference (ECAC) Institutional Strategy the previous year. A key feature of this Strategy is that “an independent Performance Review System covering all aspects of ATM in the ECAC area will be established to put greater emphasis on performance and improved cost-effectiveness, in response to objectives set at a political level”.

The PRC reviews the performance of the European ATM System under various Key Performance Areas. It proposes performance targets, assesses to what extent agreed targets and high-level objectives are met and seeks to ensure that they are achieved. The PRC/PRU ana-lyses and benchmarks the cost-effectiveness and productivity of Air Navigation Service Providers in its annual ATM cost-effectiveness (ACE) Benchmarking reports. It also produces ad hoc reports on specific subjects.

Through its reports, the PRC seeks to assist stakeholders in understanding from a global perspective why, where, when, and possibly how, ATM performance should be improved, in knowing which areas deserve special attention, and in learning from past successes and mistakes. The spirit of these reports is neither to praise nor to criticise, but to help everyone involved in effectively improving perfor-mance in the future.

The PRC holds 5 plenary meetings a year, in addition to taskforce and ad hoc meetings. The PRC also consults with stakeholders on specific subjects.

Mr. Laurent Barthelemy Ms. Marja Hutchings Mr. Marc Baumgartner General Giorgio IscraMr. Nils Billinger Mr. Antero Lahtinen Vice ChairmanMr. René Brun Ms. Anne LambertMr. Juan Bujia-Lorenzo Professor Dr. Hans-Martin NiemeierCaptain Hasan Erdurak Mr. Ralph Riedle Chairman

PRC Members must have senior professional experience of air traffic management (planning, technical, operational or economic as-pects) and/or safety or economic regulation in one or more of the following areas: government regulatory bodies, air navigation ser-vices, airports, aircraft operations, military, research and development.

Once appointed, PRC Members must act completely independently of States, national and international organisations.

The Performance Review Unit (PRU) supports the PRC and operates administratively under, but independently of, the EUROCONTROL Agency. The PRU’s e-mail address is [email protected].

The PRC can be contacted via the PRU or through its website www.eurocontrol.int/prc.

PRC PROCESSES

The PRC reviews ATM performance issues on its own initiative, at the request of the deliberating bodies of EUROCONTROL or of third parties. As already stated, it produces annual Performance Review Reports, ACE reports and ad hoc reports.

The PRC gathers relevant information, consults concerned parties, draws conclusions, and submits its reports and recommendations for decision to the Permanent Commission, through the Provisional Council. PRC publications can be found at www.eurocontrol.int/prc where copies can also be ordered.

EUROCONTROL

For any further information please contact:

Performance Review Unit, 96 Rue de la Fusée,B-1130 Brussels, Belgium

Tel: +32 2 729 3956

[email protected]://www.eurocontrol.int/prc

Performance Review Report - PRR 2014

Documents

Transcript of Performance Review Report - PRR 2014