Performance Review Report

Transcription

1 PRR 5 An Assessment of Air Traffic Management in Europe during the Calendar Year 2001 Performance Review Report July 2002 Performance Review Commission

2 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, Tel: , Fax: PRR 5 - Table of contents

3 Table of Contents EXECUTIVE SUMMARY 9 INTRODUCTION 9 TRAFFIC 9 SAFETY 9 DELAYS 10 COST-EFFECTIVENESS 11 FLIGHT EFFICIENCY 12 AIRPORTS 12 INVESTMENT IN ATM 12 FRAGMENTATION 13 CIVIL-MILITARY ISSUES 13 ATM GOVERNANCE 13 1 INTRODUCTION ABSTRACT THE PERFORMANCE REVIEW COMMISSION PRC PROCESSES 15 2 TRAFFIC INTRODUCTION TRAFFIC STATISTICS GEOGRAPHICAL TRAFFIC BREAKDOWN BREAKDOWN PER AIRCRAFT OPERATORS DEMAND FORECAST CONCLUSIONS 21 3 SAFETY INTRODUCTION ACCIDENTS SAFETY PERFORMANCE TARGETS AND ASSOCIATED INDICATORS IMPLEMENTATION OF STANDARDS AND CONFIDENTIALITY POLICY NON-PUNITIVE SAFETY REPORTING INCIDENT DETECTION, REPORTING AND ANALYSIS KEY RISK AREAS INDEPENDENT SAFETY REVIEW CONCLUSIONS 29 4 DELAYS INTRODUCTION AIR TRANSPORT DELAYS ATFM DELAYS EN-ROUTE CAPACITY AND DELAYS (EUROPEAN LEVEL) NETWORK EFFECTS (REGIONAL LEVEL) ATFM DELAYS (ANSP LEVEL) DELAY TARGETS AND CAPACITY AIRPORT ATFM DELAYS ATFM PERFORMANCE CONCLUSIONS 44 Table of contents - PRR 5-3

4 5 COST-EFFECTIVENESS INTRODUCTION KEY PERFORMANCE INDICATORS (ECAC LEVEL) UNIT COSTS PER STATE GROSS MARGIN EUROCONTROL COSTS ACTUAL 2000 COSTS COMPARED TO EARLIER PLANS BENCHMARKING OF EUROPEAN ANSPS INFORMATION DISCLOSURE FULL COST RECOVERY PRINCIPLE COST TARGET CONCLUSIONS 57 6 FLIGHT EFFICIENCY INTRODUCTION FLIGHT EFFICIENCY: INITIAL PERFORMANCE INDICATOR FUEL EFFICIENCY CONCLUSIONS 62 7 AIRPORTS INTRODUCTION AIRPORT KEY PERFORMANCE AREAS AIRPORT CAPACITY CONCLUSIONS 64 8 INVESTMENT IN ATM INTRODUCTION INVESTMENT IN ATM INVESTMENT IN R&D IMPLEMENTATION OF INDIVIDUAL ATM SYSTEMS COST OF INDIVIDUAL ATM SYSTEMS EUROPEAN PROJECTS JOINT PROJECTS MINIMUM FUNCTIONAL LEVELS AND INTEROPERABILITY ECAC INSTITUTIONAL STRATEGY CONCLUSIONS 69 9 FRAGMENTATION INTRODUCTION NATIONAL INITIATIVES REGIONAL/INTERNATIONAL INITIATIVES CONCLUSIONS CIVIL/MILITARY ISSUES INTRODUCTION MILITARY ACTIVITY AND MILITARY AIRSPACE REQUIREMENTS MILITARY REQUIREMENTS OF THE EUROPEAN ATM SYSTEM AIRSPACE - A FINITE RESOURCE ATM CIVIL-MILITARY CO-ORDINATION CONCLUSIONS PRR 5 - Table of contents

5 11 ATM GOVERNANCE INTRODUCTION THE ATM INSTITUTIONAL STRATEGY SINGLE EUROPEAN SKY ATM GOVERNANCE AT STATE LEVEL ATM GOVERNANCE AT THE EUROPEAN LEVEL CONCLUSIONS PRC RECOMMENDATIONS AND CORRESPONDING DECISIONS 85 Table of contents - PRR 5-5

6 Table of figures Figure 1: Key Performance Indicators...8 Figure 2: Traffic variations... 9 Figure 3: Adherence of EUROCONTROL States to ESARR Figure 4: Traffic and effective capacity...11 Figure 5: Rough European ATM investment breakdown...12 Figure 6: Military areas and major GAT traffic flows in the core area...13 Figure 7: Traffic indicators (2001)...17 Figure 8: Traffic variations Figure 9: Yearly and monthly traffic variation Figure 10: Evolution of passenger demand...18 Figure 11: Traffic variations from Dec to Dec Figure 12: Breakdown per traffic flows...19 Figure 13: Traffic breakdown per States...19 Figure 14: Distribution of traffic per aircraft operators...20 Figure 15: Breakdown per type of operators...20 Figure 16: Actual traffic vs. forecast...21 Figure 17: Fatal accidents in geographical Europe in 2001 (commercial traffic)...23 Figure 18: Accident categorisation (IFR and VFR)...24 Figure 19: Incidents reported by ECAC States...25 Figure 20: Safety targets and indicators...26 Figure 21: Adherence of EUROCONTROL States to ESARR Figure 22: Implementation of ICAO Annex 13, para in national law...27 Figure 23: Barriers to reporting...27 Figure 24: Air Transport punctuality...31 Figure 25: Air transport departure delay causes (excluding reactionary)...32 Figure 26: Traffic and en-route ATFM Delays (2000, 2001)...33 Figure 27: Summer ATFM Delays...33 Figure 28: En-route and airport ATFM delays Figure 29: Key European traffic and delay data...34 Figure 30: Traffic and effective en-route capacity...34 Figure 31: En-route ATFM delay costs in Figure 32: Effective en-route capacity and traffic...35 Figure 33: ATFM En-route delays in European regions...36 Figure 34: ATFM delays in European regions...37 Figure 35: Traffic, actual and expected ATFM en-route delay (2001)...37 Figure 36: En-route delays in Rhein and Maastricht UAC...38 Figure 37: Summer ATFM en-route delay per ANSP...38 Figure 38: Summer en-route ATFM delay target...39 Figure 39: Required and committed level of capacity...41 Figure 40: Airport ATFM delays (2001)...42 Figure 41: Airport ATFM delay causes...42 Figure 42: Over-deliveries...43 Figure 43: Non-adherence to ATFM slots...43 Figure 44: Top airports and airlines in non-adherence reports...44 Figure 45: Evolution of en-route costs...47 Figure 46: Evolution of real unit cost, total costs and traffic...47 Figure 47: Unit costs per State...48 Figure 48: Change in unit rates between 2001 and PRR 5 - Table of contents

7 Figure 49: Cost breakdown by type (2000 Data) Figure 50: National costs breakdown by type (2000 data) Figure 51: Evolution of EUROCONTROL Agency costs (Part I)...51 Figure 52: Breakdown of Agency costs per establishment Figure 53: Evolution of EUROCONTROL Maastricht costs...52 Figure 54: Evolution of CRCO costs...52 Figure 55: planned costs compared to actual...53 Figure 56: Actual versus estimated unit cost per km (2000 data)...54 Figure 57: Unit cost trend (qualitative)...57 Figure 58: An example of ARTAS recorded tracks...59 Figure 59: Flight length and horizontal flight inefficiency distribution...60 Figure 60: Flights between Northern Italy and Benelux...61 Figure 61: Airport operations in the gate-to-gate ATM...63 Figure 62: Example of initial, allocated and actual demand...64 Figure 63: Rough European ATM investment breakdown...65 Figure 64: European ATM R&D funding...65 Figure 65: ARTAS implementation map...67 Figure 66: Traffic and size of European and US en-route centres...71 Figure 67: Reduction of en-route centres in Australia...73 Figure 68: Airservices staff evolution...74 Figure 69: Variation of military and GAT traffic...77 Figure 70: Military areas and major GAT traffic flows in the core area...78 Figure 71: Common Performance Model...81 Table of contents - PRR 5-7

8 European ATM Performance States Traffic Forecast Actual IFR flights M(+5.3%) M(-0.6%) Source: CFMU Distance Source: CRCO Safety M km (+0.2%) KPI Accidents with direct ATM contributions Incidents Upper limit Not applicable To be set Actual 1 Not accurately known Delays KPI: Average summer en-route ATFM delay Minutes per Flight ,9 4,1 5,5 3,6 3,1 Actual : 3.1 min/flight Target : 2.8 min/flight Summer A ATFM delay: 27.6 M min En-route delay cost: M EUR Optimum delay Target delay Actual delay Data Source: CFMU 0,9 Cost-Effectiveness KPI: Real unit cost 260 0,8 0, Present Trend Euro 2000 / km 0,6 0, est 2002 est Cost Per Km Total Costs (1990= index 100) Distance (1990=index 100) All states in CRCO system Data source :CRCO Figure 1: Key Performance Indicators 8 - PRR 5 - Executive summary Charges Route: 4800 M T Terminal: ~1400 M T Total: ~6200 M

9 EXECUTIVE SUMMARY Introduction This fifth Performance Review Report (PRR 5), covering the year 2001, assesses first the performance of the European Air Traffic Management (ATM) system under four Key Performance Areas: Safety; Delays (and capacity); Cost-effectiveness; and, Flight efficiency. The report shows that there were deficiencies in all these areas and indicates some of the improvements that would be needed: airports, investment in ATM, fragmentation, civil-military co-ordination and ATM governance. The European Union, EUROCONTROL, and States have an opportunity to address these issues in the context of the Single European Sky, expected to be in place by 31 December Traffic As shown in Figure 2, air traffic growth will have virtually stopped in 2001 (-0.6% controlled flights) and 2002, after several years of strong increases. After 11 September, lower passenger demand (some -10% in Europe) resulted in severe financial difficulties for many airlines, lower traffic demand, lower revenues for Air Navigation Service Providers (ANSPs), and lower delays. In this context, attention is likely to shift temporarily from delays to costs, with the risk that fewer resources are directed towards increasing capacity. In similar past events, such as the Gulf War, traffic demand quickly recovered after a temporary down-turn. Reduced traffic forecasts and lower pressure on capacity exposed the ATM system to high delays in subsequent years. Underestimating future traffic growth in the wake of specific events has proved to be a mistake. Safety Figure 2: Traffic variations ATM had a direct contribution to the accident which occurred on a Paris-CDG runway in May There is a worrying recurrence of aircraft collisions on runways (Milan Linate, 2001), which calls for full support being given to current initiatives on runway incursions. The accident in Zürich (November 2001) raises the question of safety given specific environmental decision-making. Executive summary - PRR 5-9

10 In the absence of adequate performance indicators, safety trends and the extent to which the safety objectives of the ATM Strategy are being met are unknown. European Civil Aviation Conference (ECAC) safety objectives on incidents and accidents should be quantified and incorporated in the revised ATM Strategy. Appropriate indicators should be published annually to show if agreed targets are being achieved. As shown in Figure 3, a number of EUROCONTROL Member States have not fully complied with the Commission Decision No. 80 (1999) mandating implementation of EUROCONTROL Safety Regulatory Requirements (ESARR 2). This raises questions about the effectiveness of EUROCONTROL decisions. Insufficient compliance of States with the EUROCONTROL safety reporting standard ESARR 2 is a major obstacle to measuring matching indicators, identifying key risk areas and managing safety. ESARR 2 Phases Phase 1: Due by 1 Jan 2000 Reporting of accidents and incidents-near collisions Phase 2: Due by 1 Jan 2001 Reporting of incidents with a potential to become collisions or near collisions - Source: SRC Phase 3: Due by 1 Jan 2002 Reporting of ATM specific occurrences having an impact on the ability to provide safe ATM Figure 3: Adherence of EUROCONTROL States to ESARR 2 Currently, an overly restrictive interpretation of the EUROCONTROL Publication and Confidentiality Policy prevents the naming of non-complying States. This policy should apply only to the protection of individuals and detailed safety data, not to States compliance with safety processes and EUROCONTROL standards. In some States, legislation hinders penalty-free incident reporting. This is contrary to ICAO Annex 13 and European Commission (EC) Directive 94/56/EC, and is an obstacle to the flow of incident data. As long as many incidents remain undetected, it will be difficult to determine the exact risk level and its trend. Legislation providing at least minimum legal protection for individuals reporting safety occurrences is needed in all States. Each State should ensure the creation of safety review functions independent of other national structures. Experience should be shared among States and with EUROCONTROL, and disseminated as appropriate within the aviation community. Delays Although traffic grew by only +0.2% during the Summer 2001 period (May to October), well below planned growth (+5.3%), en-route Air Traffic Flow Management(ATFM) delays (3.1 min) were above the interim target (2.8 min), as shown in Figure 1. Had traffic grown as expected, delays would have been even higher (an estimated 35%). As shown in Figure 4, capacity gains in 2001 were substantial but came too late to have a strong influence on summer delays. The year 2002 is expected to be better, due to relatively low levels of traffic and the positive effects of airspace changes and system implementations PRR 5 - Executive summary

11 Flights Kosovo ARN V3/3 NERC Jan 1997 Jul 1997 Jan 1998 Jul 1998 Jan 1999 Jul 1999 Jan 2000 Jul 2000 Jan 2001 Jul 2001 Jan 2002 Daily traffic Month Effective capacity Figure 4: Traffic and effective capacity According to the ATM Strategy, capacity should meet demand under normal circumstances. Effective capacity has increased, but lagged behind demand over recent years. This indicates a lack of responsiveness within the European ATM system under the present institutional and incentive arrangements. The cost of ATFM en-route delays to airspace users was in the order of 1100 to 1700 M in ATFM delays were mostly due to insufficient Air Traffic Control (ATC) capacity, particularly in Area North and London in the high traffic season. European ATM network inefficiencies and staffing issues appear to be the most critical factors behind these delays. There should be no reduction in effort to increase capacity to meet and maintain the optimum en-route ATFM delay target (1 minute per flight). Action plans to this end should be incorporated in relevant ANSP business plans. It is not just a matter of ATC capacity; ATFM is not using existing capacity to best effect. Some 25% of ATFM delays are due to unnecessary ATFM regulations and lost slots. In addition, the ATFM system does not fully protect sectors against over-deliveries, leading to understated declared capacities. This is largely due to abuse of ATFM slots, ATFM techniques such as re-routing remaining under-used, and the fact that the Central Flow Management Unit (CFMU) has no clear mandate for tactical flow management at European level. Binding rules of application for ATFM and a review of ATFM principles and operating modes are needed. This must be addressed urgently as part of the ATFM action plan. Cost-effectiveness Lower traffic growth has had a negative impact on unit costs. In this context, ensuring cost-effective ATM is of the utmost importance. The recent upward trend in en-route unit costs (see Figure 1) is raising concerns. The current cost recovery system does not provide incentives for ANSPs to deliver the right level of capacity and to control their costs. It is also unable to cope adequately with a sudden change in traffic. Principles and application of alternative charging mechanisms should be investigated with urgency, together with other measures to improve ANSP flexibility and responsiveness. To ensure a balanced approach, it is important that the Provisional Council sets European-wide objectives for safety, delays and unit costs. It is even more important that each ANSP defines and is committed to clear performance targets. A significant improvement in the target setting process needs to be introduced both at European and national levels. The Performance Review Commission (PRC) will propose cost performance targets for the European ATM system to the Provisional Council. Executive summary - PRR 5-11

12 PRR 4 showed that the unit cost of the European ATM system (en-route and terminal) is some 70% above its US counterpart. From this and benchmarking within Europe, the cost-effectiveness issue appears to be of the same order of magnitude as the delay issue. While this conclusion must be treated with great caution, it indicates potential considerable gains if effective decisions are taken on Single European Sky issues. The PRC has drawn up, and the States have agreed, a Specification for Information Disclosure. This is enabling benchmarking, an essential tool for ANSP management, ATM regulation and performance review. The experience gained is shared world-wide with interested States/ANSPs, with the aim of achieving maximum commonality of reporting and performance analysis. Flight efficiency Initial data indicate that average horizontal flight inefficiency (route extension) for the core area of Europe varies between 10-12% for short flights (under 400km) and 8% for routes between 400km and 1000km. Flight inefficiency has an influence on the environment (fuel burn, emissions), which remains to be assessed. Airports The traffic downturn offers an opportunity for en-route capacity to match demand earlier than foreseen. If this is achieved, airport issues will become dominant, as is the case in the United States of America (US). The effort devoted to airport ATM-related issues has so far not been commensurate with the issue at hand, nor with the gate-to-gate strategy. The PRC is planning to conduct an airport survey with special emphasis on airport capacity and its utilisation. It will also review terminal charges (approximately 22% of total ATM costs). This survey had to be postponed to 2003 due to budgetary reductions. Investment in ATM Approximately 600 M are invested in European ATM systems and Research and Development (R&D) every year as shown in Figure 5. This is no more than 10% of ATM costs, which is low by industry standards. Total ATM systems R&D Other (CNS, infrastructure) 1200 M 400 M 200 M 600 M Figure 5: Rough European ATM investment breakdown Nearly all tailor-made ATM system projects in recent years have suffered from major delays (up to five years), cost overruns and reduced functionality. The new ATM systems should be based on standard industry products to the maximum extent possible. Financial transparency of investments is poor. Information disclosure requirements should be extended to include data on capital expenditure broken down by major investments (including e.g. investment in ATM systems, Communications, Navigation, Surveillance (CNS), R&D and infrastructure projects). ANSPs and aircraft operators should be bound to meet European minimum functional requirements and interoperability standards by specified dates, according to an agreed Master Plan. This would ensure synchronised implementation and encourage development of compliant products by industry. A central European body should facilitate the adoption of an overall system architecture and of a Master Plan. European ATM R&D is very fragmented and has not led to any major breakthroughs in ATM operations and technology in the last 30 years. ATM R&D should be organised much more efficiently PRR 5 - Executive summary

13 Fragmentation Fragmentation has been identified in PRR 4 as one of the reasons for the cost-effectiveness and productivity performance gaps between the US and European ATM systems. Fragmentation covers a wide range of issues: fragmentation of airspace, ANSPs, Air Traffic Service (ATS) units, civil and military ANS provision, regulation, ATM systems and interfaces. The PRC is studying this, and results will be published in PRR 6. Several initiatives aimed at reducing civil-military, organisational and airspace fragmentation have proved that operational and economic efficiency improvements can be made while maintaining safety. Existing and new initiatives should be encouraged. Corresponding provisions, which already exist under International Civil Aviation Organisation (ICAO) policy (ATS delegation), have been expanded by EUROCONTROL for application in Europe, and could be further reinforced by the European Union (EU) Single European Sky legislation when adopted. Civil-Military issues ATM civil-military co-ordination in the European core area has a strong influence on civil traffic flows and military activities. Figure 6 shows the interaction between the major General Air Traffic (GAT) flows and the Temporary Segregated Areas (TSAs) in core Europe Figure 6: Military areas and major GAT traffic flows in the core area Efficient civil-military co-ordination is needed within States and across borders. This is a pre-requisite for the success of the Single European Sky. At the moment there is no process to ensure the cross-border compatibility of national civil-military airspace designs. This penalises the European airspace planning process. The harmonisation of operational air traffi c (OAT) rules at European level would signifi cantly improve the handling of interactions between GAT and OAT. Poor dissemination of airspace status information results in a signifi cant amount of airspace not being used. A single European source of airspace status information would significantly increase the efficiency of airspace management and use. There is a need and room to reduce signifi cantly the amount of temporary segregated airspace for the benefi t of both civil and military users. In the short-term, this consists of managing the airspace more efficiently. In the medium term, a more efficient and integrated (at least operationally) organisation of ATS would enable some types of military activities to be performed without airspace segregation. The PRC endorses the nine recommendations contained in the report on the status of civil-military coordination in Europe and awaits results and a calendar of implementation dates. Data source: CFMU & AIP Executive summary - PRR 5-13

14 ATM governance National ATM governance arrangements have a key role to play in ensuring ATM performance targets are met. Governance arrangements comprise the roles of the regulatory authorities and responsibilities of the owner (usually the State) and its ANSP. The PRC encourages the application of five ATM governance principles: 1.Clear accountability of the ANSP to the owners and regulators; 2.Clear high level ANSP performance objectives endorsed by the State (safety, capacity/delay, cost-effectiveness); 3.Explicit statements about autonomy of the ANSP in pursuing the performance objectives; 4.Incentives for the ANSPs to achieve the performance objectives; 5.Definition of rights and roles of regulators, owners and stakeholders at the national and European levels. States (and/or groups of States) should require their ANSPs, in their business plans, to publish clear objectives (safety, capacity/delays, costs) and commit to meeting them after consultation with airspace users. ANSPs have a prime responsibility to work closely with each other so that an effective European ATM network is in place. EUROCONTROL should facilitate this and be accountable within the European regulatory framework PRR 5 - Executive summary

15 1 Introduction 1.1 Abstract The fifth Performance Review Report (PRR 5) of EUROCONTROL s Performance Review Commission (PRC) presents: Traffic demand placed on the European Air Traffic Management (ATM) system in 2001, including influence of 11 September events (chapter 2); Its response under four Key Performance Areas, i.e. safety (chapter 3), delays (chapter 4), cost-effectiveness (chapter 5), flight-efficiency and related influence on environment (chapter 6, new in PRR 5). PRC analysis of some critical elements for ATM performance: airports (chapter 7), investment in ATM systems (chapter 8), fragmentation (chapter 9), civil-military issues (chapter 10) and ATM Governance (chapter 11). PRC recommendations to the Provisional Council (PC) (developed after the Consultation Meeting), and corresponding decisions (added after PC 14). 1.2 The Performance Review Commission The PRC was established in 1998 by the Permanent Commission of EUROCONTROL. In doing so, the Permanent Commission implemented a recommendation of the European Civil Aviation Conference (ECAC) Institutional Strategy [ref.1] which states 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 is composed of 12 independent Commissioners. They are selected on the basis of ability, competence, experience and professional reputation in the field of aviation The PRC advises the Permanent Commission through the Provisional Council on all matters relating to the development of a strong, transparent and independent performance review and target-setting system. This includes all aspects of air traffic management, including policy and planning, airport and airspace safety management, and financial and economic aspects of services provided. The PRC is supported by the Performance Review Unit (PRU) More information about the PRC is provided on the inside-back cover of this report. 1.3 PRC processes The PRC reviews ATM performance issues on its own initiative, at the request of the governing bodies of EUROCONTROL or of third parties. It prepares annual and ad hoc Performance Review Reports (PRR), and other reports to the Permanent Commission, through the Provisional Council, as required. It assembles relevant information, consults concerned parties and draws conclusions. These conclusions, after consultation in open session with interested parties, are used to develop recommendations to be submitted for decision to the Permanent Commission, through the Provisional Council. The PRC monitors actions taken following its recommendations [ref.2] The underlying methodology used in PRRs, e.g. Key Performance Areas (KPAs) and Indicators (KPI) is developed in consultation with interested parties [ref.3]. The principal KPAs used to date are safety, delays and cost-effectiveness. For the safety KPA, the PRC uses safety data provided by the Safety Regulation Commission (SRC) and other sources. The PRC s conclusions are checked with the SRC s supporting unit - the Safety Regulation Unit (SRU) PRC documents and reports are available from the PRC web site ( or from the PRU (pru@eurocontrol.int). Introduction - PRR 5-15

16 16 - PRR 5 - Introduction

17 2 Traffic 2.1 Introduction This chapter characterises GAT IFR 1 traffic demand and its variations. The response of the ATM system to the traffic shock following the economic slow-down and 11 September events is analysed in the following chapters. 2.2 Traffic statistics Key indicators for GAT IFR traffic are the number of flights and the distance flown, the latter being a better measure of services provided, and directly related to route charges. Figure 7 gives key traffic indicators for Figure 7: Traffic indicators (2001) Variations of both indicators in the period are summarised in Figure 8. Approximate growth from 1990 to 2001 Average yearly growth See Number of IFR flights 60% 2 4.8% Figure 9 Distance flown in a constant area 80% 3 6% Figure 30 Figure 8: Traffic variations Variations in the distance flown are the EUROCONTROL area is driven by variations in three factors: number of flights, average distance flown, size of the area. The combination of the three factors resulted in distance flown increasing by 140% from 1990 to 2001 in the (expanding) EUROCONTROL area (see bottom part of Figure 1). Figure 9: Yearly and monthly traffic variation 4 Traffic - PRR 5-17

18 2.2.4 Figure 9 shows the monthly evolution of flights in the EURO99 area. The year on year increase in traffic was low during the first eight months of the year reflecting the economic slowdown. A net decrease in traffic can be observed after 11 September As can be seen in Figure 10, passenger demand decreased sharply after 11 September. However, it took some time for airlines to adjust their schedules accordingly, the number of flights decreasing only slightly in September and October (see Figure 9). The introduction of the winter schedule and the collapse of some major European airlines, resulted in significant traffic reduction in November and December. Traffic in December (-9.7% compared to December 2000) dropped below 1998 levels. Figure 10: Evolution of passenger demand Traffic variation differs widely across countries, as shown in Figure 11. total flights (including overflights) Departure per country Dec 2001 compared to Dec 2000 More than 2.5% 2.5% to -2.5% -2.5% to -7.5% -7.5% to -12.5% -12.5% to -17.5% -17.5% and less Figure 11: Traffic variations from Dec to Dec Source: STATFOR 18 - PRR 5 - Traffic

19 2.3 Geographical traffic breakdown As shown in Figure 12, domestic flights (departing and arriving within the same country) represent 39% of flights in Europe. 47% of flights are intra-european (i.e. airports of departure and arrival are within two different European countries). Only 14% of flights are extra-european. During the last five years the number of domestic flights has increased by 12%, intra-european flights by 29% and extra-european flights by 36% Extra European 13.8 Domestics Intra European Figure 12: Breakdown per traffic flows Data source: CFMU Figure 13 represents a breakdown per States. The proportion of over-flights, departure/arrival international flights, and domestic flights vary significantly from State to State. Figure 13: Traffic breakdown per States Traffic - PRR 5-19

20 2.4 Breakdown per aircraft operators Figure 14 shows that the top 10 airlines operate 25% of European flights, with 75% operated by the top 100 carriers. 100% Percentage of flight 75% 50% 25% 0% 75% of the flights are operated by the top 100 airlines Number of aircraft operators Figure 14: Distribution of traffic per aircraft operators Figure 15 shows the distribution of traffic per type of operators. National flag-carriers operate the majority of the flights. Most of those airlines are members of the Association of European Airlines (AEA). Two of them, SABENA and Swissair, did not survive the year The consolidation process of the European airline industry also resulted in a number of regional airlines disappearing or being integrated under the umbrella of major flag-carriers. Data source: CFMU Figure 15: Breakdown per type of operators Data source: CFMU In contrast, the low cost segment continued to experience rapid growth in 2001 (+41%). However, it remains a very small part of the total number of flights (4%) Traffic patterns evolve as airlines adjust continuously to market forces, driven by passenger preferences and liberalisation. The European ATM system should be sufficiently flexible to respond to changes in traffic patterns PRR 5 - Traffic

21 2.5 Demand forecast Figure 16 compares the actual traffic (blue line) with the forecast made three years earlier (orange line). Although traffic growth (in terms of flights) stopped in 2001, the traffic recorded in 2001 is still within the bounds of the forecast published in June STATFOR forecasts have been revised downwards significantly. Compared with the June 2000 forecast which anticipated a 19% increase in traffic between (top of the orange line), the June 2001 forecast already indicated a two-year shift in demand (i.e. +19% achieved only in 2005) According to the latest forecast published by STATFOR in February 2002 (dotted line), the 19% increase originally foreseen in 2003 will now only be reached in The strong traffic growth over the last seven years will have virtually stopped for two years in 2001 and Past experience shows, however, that forecasts tend to be biased by the perception of the current situation, resulting in an overestimation of future demand when traffic growth is high and an underestimation in periods of low traffic growth. Figure 16 shows that, following the low traffic increase recorded in 1993 (blue line), the forecast for was revised downward. Traffic however caught up rapidly and delays increased significantly. Figure 16: Actual traffic vs. forecast Data source: STATFOR 2.6 Conclusions Air traffic growth will have virtually stopped in 2001 (-0.6% controlled flights) and probably 2002, after several years of strong increase. After 11 September, lower passenger demand (some -10% in Europe) resulted in severe financial difficulties for many airlines, lower traffic demand, lower revenues for ANSPs, and lower delays. In this context, attention is likely to shift temporarily from delays to costs reduction, resulting in fewer resources being directed towards increasing capacity In similar past events, such as the Gulf War, traffic demand quickly recovered after a temporary down-turn. Reduced traffic forecasts and lower pressure on capacity exposed the ATM system to high delays in subsequent years. Underestimating future traffic growth in the wake of specific events has proved to be a mistake. Traffic - PRR 5-21

22 Notes on chapter 2 1 General Air Traffic (GAT) flying under Instrument Flight Rules (IFR). 2 Within the EURO88 area, composed of airspace controlled by 11 States (see glossary), which corresponds approximately to 75% of traffic in Cumulated yearly variations within a constant area. 4 The variation of the number of flights was 0.6% in CFMU area (source CFMU) and -0.4% in EURO88 area (source STATFOR) PRR 5 - Traffic

23 3 Safety 3.1 Introduction Safety is of the highest priority in aviation. Accordingly, it was identified as the first Key Performance Area by the PRC from its inception in Since then, the PRC has consistently advocated 5 the need for a transparent system in which all relevant safety occurrences are reported and analysed so that problems can be identified and resolved at an early stage before they result in loss of life, injury, or material damage. 3.2 Accidents ATM is generally considered to be very safe, but there is a lack of hard evidence to identify potential risks. There have been four fatal accidents with a direct ATM contribution 6 involving commercial aircraft in the last 25 years (1976 to 2000) 7 : Zagreb, 1976; Tenerife, 1977; Madrid, 1983; Paris CDG, 2000 [ref.4] Nine accidents involving commercial aircraft occurred in geographical Europe in 2001, causing 274 fatalities, two of which were in core Europe (see Figure 17). Airline Aircraft Date/Place Fatalities Loganair Shorts Edinburgh 2 Russ Air Transport Il-76TD Moscow 10 Binter Mediterràneo CASA Màlaga 4 Sibir Airlines Tu Black Sea 78 SAS MD Milan-Linate Flightline Metro Swearingen Mediterranean Sea, off Valencia IRS Aero Il Kalayzin 27 ELK Airways AN Kärdla, Estonia 1 Crossair RJ Zurich 24 Total Source: Aviation Safety Network Figure 17: Fatal accidents in geographical Europe in 2001 (commercial traffic) The remainder of sections 3.2 and 3.3 is based on Annual Summary Templates (AST) provided by States to the SRU in fulfilment of ESARR 2, [ref.5.] The latest AST include 1999 data, initial 2000 data but no 2001 data 9. Safety - PRR 5-23

24 TYPE OF REPORTED ACCIDENT (fatal accidents in brackets) Final 1999 Initial 2000 Total number of mid-air collisions 14 (7) 9 (3) Total number of CFIT 71 (21) 12 (3) Total number of collisions on the ground between a/c 3 (0) 9 (0) 10 Total number of collisions between an airborne aircraft and vehicle/ another aircraft on the ground Total number of collisions on the ground between aircraft and vehicle /persons/ obstruction(s) Figure 18: Accident categorisation (IFR and VFR) (1) 97 (0) Source: SRC DOC 2 [ref.6] From the analysis of accident data (see Figure 18 and Annex 1 for other statistics), the SRC drew the following main conclusions: None of the accidents were declared as having an ATM DIRECT contribution (in available reports); Three accidents involving commercial IFR flights (of which one was fatal), were identified as having an ATM INDIRECT contribution; Of the reported mid-air collisions in 2000, there is no indication of IFR/IFR implication, and only in one case was ATM identified as having an INDIRECT contribution; Of the reported CFIT incidents in 2000, five were declared as having an IFR implication, and only two (of which one was fatal) were declared as having an ATM INDIRECT contribution. The significant increase in the reported number of collisions on the ground between aircraft and vehicle(s)/ person(s)/obstruction(s), which had no declared ATM contribution, needs further analysis Key risk areas, in particular runway incursions, are discussed in section PRR 5 - Safety

25 TYPE OF INCIDENTS REPORTED INCIDENTS 11 (reporting levels and States differ for both years) Final 1999 Initial 2000 Separation minima infringement Inadequate separation Near Controlled Flight Into Terrain (CFIT) 6 1 Runway Incursion Runway excursion by aircraft 2 45 Aircraft deviation from ATM clearance Aircraft deviations from applicable Published ATM procedures Unauthorised penetration of airspace Source: SRC Annual Report Figure 19: Incidents reported by ECAC States The data set in Figure 19 is largely inconsistent and incomplete. The SRC was however able to draw some conclusions (see also Annex 1), the following being particularly important: In 2000 the number of AIRPROX reports was reduced compared to 1999, but still higher than in The number of reports involving OAT traffic seems disproportionate to the amount of OAT traffic, which might be an indication of a potential key risk area in civil/military co-ordination Inconsistencies and gaps in ASTs should be corrected as soon as possible by the responsible organisations and all States should report (see section 3.5). 3.4 Safety performance targets and associated indicators The ATM Strategy safety objective is to improve the safety level by ensuring that the number of ATM induced accidents and serious or risk-bearing incidents do not increase and, where possible, decrease In the absence of quantified safety objectives 13 and of corresponding performance indicators, it is not clear whether the ECAC safety objective is being met or not In order to avoid accidents, it is essential to obtain reliable information on incidents. Unfortunately, the information received from States is not yet adequate for the SRC to publish reliable Key Performance Indicators and safety targets. As long as data remain inconsistent and incomplete, it will be difficult to assess if targets are being met. Safety - PRR 5-25

26 3.4.4 ECAC safety objectives on incidents and accidents should be quantified and incorporated in the revised ATM Strategy [ref.7]. Appropriate indicators should be published annually to show if agreed targets are being achieved (see Figure 20). Figure 20: Safety targets and indicators States also should set national targets, as recommended in the SRC Policy DOC 1 [ref.8] and amendment 40 to ICAO Annex 11. EUROCONTROL presently has no evidence of whether this is being done. 3.5 Implementation of standards and confidentiality policy Phase 1 of ESARR 2 is applicable in all Member States from 1 January 2000 and Phase 2 from 1 January However, ESARR 2 has been implemented to variable levels 14 or not at all by some EURO- CONTROL Member States (see Figure 21). This is contrary to the EUROCONTROL Permanent Commission s Decision No. 80 dated (further supported by Transport Ministers at MATSE 6). ESARR 2 Phases Figure 21: Adherence of EUROCONTROL States to ESARR 2 Source: SRC Phase 1: Due by 1 Jan 2000 Reporting of accidents and incidents-near collisions Phase 2: Due by 1 Jan 2001 Reporting of incidents with a potential to become collisions or near collisions - Phase 3: Due by 1 Jan 2002 Reporting of ATM specific occurrences having an impact on the ability to provide safe ATM All States must be able to demonstrate their conformance to international and European requirements and standards. There is no point in setting standards if their application cannot be verified. However, the EUROCONTROL Publication and Confidentiality Policy [ref.9] has been interpreted so as to prevent the publication of the list of non-compliant States The EUROCONTROL Publication and Confidentiality Policy should apply only to the protection of individuals and detailed safety data, not to the States compliance with safety processes and EUROCONTROL standards PRR 5 - Safety

27 Source: SQS Unit 3.6 Non-punitive safety reporting As announced in PRR 4 [ref.10], the PRU has launched a safety survey among EUROCONTROL States to assess the implementation of a non-punitive legislative environment and a blame-free safety culture. These have been encouraged by the EC s High Level Group [ref ] and the Provisional Council Paragraph 5.12 of ICAO Annex 13 limits the disclosure of records relating to an accident or incident investigation to cases determined by a State s appropriate authority for the administration of justice. However, one third of the PRU questionnaire respondents indicated that this did not form part of national law and half of those have not filed any difference with ICAO despite an obligation to do so (see Figure 22). ICAO 18 further warns against the inappropriate use of those records for subsequent disciplinary, administrative, civil and criminal proceedings, as this could hamper flight safety. Figure 22: Implementation of ICAO Annex 13, para in national law Source: PRU A clear distinction should be made between the concept of non-punitive safety reporting, where an individual should not be sanctioned or be subject to legal proceedings if he/she voluntarily reports an incident, and the potential responsibility and liability in case of gross negligence Most of the States who have responded to the survey offer neither total nor partial immunity to those voluntarily reporting safety occurrences, and the initial analysis shows that, usually, in the context of judicial enquiries, the authority for the administration of justice can have access to the evidence and results of a technical investigation These issues are being further studied. More details will be available in the survey s final report (July 2002) and in PRR As long as many incidents remain undetected, an unknown proportion of risk remains invisible. Legislation providing at least minimum legal protection for individuals reporting safety occurrences is needed. 3.7 Incident detection, reporting and analysis Figure 23: Barriers to reporting The PRC has already advocated the extension of systematic detection and analysis of incidents, already performed routinely in some States, to all ECAC States Low-level reporting and lack of trust, due in part to penalties, blame, confidentiality, etc. leads to insufficient data being made available to safety managers (see Figure 23 and ref.12). This could be a major issue throughout the safety data path from reporting through to analysis and the dissemination of results. Safety - PRR 5-27

28 3.7.3 Within the airline industry a number of airlines have voluntarily participated in a systematic analysis of their Digital Flight Data Recorders (DFDRs) as part of the Flight Operational Quality Assurance (FOQA) programme. To support this and as a response to some data confidentiality issues, the FAA has recently codified regulations to provide protection against disclosure of certain safety information and security information submitted to the Federal Aviation Administration (FAA) on a voluntary basis The PRC considers that ANSPs should perform similar systematic analyses of recorded information. Moreover, work should continue on the systematic detection of all ATM operational and technical occurrences with potential safety implications, followed by immediate investigation and corrective actions. These processes (i.e. detection and analysis) should be supported to the appropriate extent by automation. However, automated tools should not prevent, or replace human reporting, they should complement it 20. A wealth of information would then become available to ANSPs to support accident prevention. 3.8 Key risk areas Full visibility and exhaustive analysis of incidents will allow key risk areas to be established and prevention programmes put in place, thereby reducing the risk of accidents. However, as long as some States fall short of a full-scale implementation of ESARR 2, there can only be general indications as to key risk areas. Some of those are detailed below Last year was marked by the second worst accident in more than a decade (in terms of total fatalities) within ECAC where an aircraft collided with another on a runway 21 (Milan Linate, 08 Oct 2001, 118 fatalities). Another fatal accident with similar characteristics occurred the previous year (Paris CDG, 25 May 2000, one fatality). The number of runway incursion incidents reported to the SRC has also increased, according to reported data and the SRC s conclusions (see Figure 18 and Figure 19) In the US, the National Transportation Safety Board (NTSB) has placed runway incursions on the most wanted list for more than a decade, consistently demanding risk-reduction actions. The FAA has announced 22 that, as a consequence of vigorous mitigation actions, runway incursions have decreased in 2001 for the first time in the previous five years While it is true that airport operations in the US differ from those in Europe, there is enough evidence to conclude that runway incursions deserve special attention in Europe. Full support should be given to the runway safety initiative On the night of 24 November 2001, a Crossair flight crashed in Zurich a few minutes after 22:00 hrs. The crew had to switch to the only runway allowed at that time and to fly a VOR/DME non-precision approach. This type of procedure is known to be more demanding than a precision approach (according to the Flight Safety Foundation). In this case, the use of a non-precision approach to accommodate environmental restrictions might have affected safety [ref.13 & 14] Two other CFIT accidents have occurred at the same airport during the past decade, ATM being identified as an indirect cause in one case: The approach controller did not observe the aircraft leaving the cleared altitude 23. Despite a decrease in reported occurrences in 2000, CFIT remains a significant hazard, accounting for roughly one third of all accidents and fatalities [ref.6]. ATM/CNS could assist flights more effectively to prevent such accidents 24, the Joint Aviation Authorities (JAA) already consider that ATM can play a role One or more States has notified problems with issues such as call-sign confusion 25, dual use of English and national languages 26, level-busts 27 and penetration of segregated airspace 28 (e.g. civil/military, IFR/VFR). Whether these are local or general issues is not clearly known The SRC (as well as ANSPs, national safety regulators and the Agency) is presently unable objectively to assess the seriousness of specific key risk areas due to the partial implementation of ESARR PRR 5 - Safety

29 3.9 Independent Safety Review Safety processes in each ATM organisation should be in accordance with EUROCONTROL requirements and ICAO SARPs The newly extended ICAO Universal Safety Oversight Audit Programme (IUSOAP) is expected to ensure that States have implemented the necessary regulatory measures to verify that ICAO safety processes are implemented in each ANSP. The SRC could possibly monitor effective and consistent implementation of ESARRs at national level in a similar manner ATM safety occurrences should be analysed and the related recommendations widely disseminated through a safety review function at national level independent of other national structures (e.g. ANSP, regulator, airline or airport operators). Experience should be shared among States and with EUROCONTROL, and disseminated as appropriate within the aviation community Conclusions ECAC safety objectives on incidents and accidents should be quantified and incorporated in the revised ATM Strategy. Appropriate indicators should be published annually to show if agreed targets are being achieved A number of States have not yet fully complied with the EUROCONTROL Commission Decision No. 80 (1999) mandating implementation of ESARR The EUROCONTROL Publication and Confidentiality Policy should apply only to the protection of individuals and detailed safety data, not to States compliance with safety processes and EUROCONTROL standards As long as many incidents remain undetected, an unknown proportion of risk remains invisible. Legislation providing at least minimum legal protection for individuals reporting safety occurrences is needed Each State should ensure the creation of safety review functions independent of other national structures. Experience should be shared among States and with EUROCONTROL, and disseminated as appropriate within the aviation community There is a worrying recurrence of aircraft collisions on runways, which calls for full support to be given to the runway safety initiative. Safety - PRR 5-29

30 Notes on chapter 3 5 Most of the safety conclusions drawn in previous PRC Reports are still valid, as the same limitations apply to the year 2000 safety data reported to the SRC. Significant variations still exist in the scope, depth, consistency and availability of ATM safety data. A summary of those recommendations and their follow-up can be found in ref.2. 6 Terminology is defined in SRC documentation (HEIDI taxonomy). 7 Source: Aviation Safety Network and Jet Airliner Crash Data Evaluation Centre (JACDEC). 8 The accident caused 110 fatalities in the SAS flight, four in the other aircraft and four on the ground. 9 The data reported by States represent the results of national investigations. Time to analyse and draw accurate conclusions can be significant. 10 The Paris-CDG (May 2000) accident had a direct ATM contribution and caused one fatality. It is not included in SRC s Annual Report 2001, as the final accident report was not available at the time of publication. 11 Differences between the present table and that published in PRR 4 (Table 4, pg. 12) for the year 1999 are due to additional reports arriving after PRR 4 publication and/or further refinements of 1999 data sent together with initial data for the year EUROCONTROL Provisional Council paper PC/01/12/14, 08/11/ A target level of safety for commercial air transport (0.623 accident per year with a direct ATM contribution [ref.6]) has been defined by the SRC mainly for design purposes, and for safety performance monitoring in the long run. 14 In addition, it is concluded that the involvement of national accident investigators, ATM safety regulators and ATM service providers in the implementation of ESARR 2 is still too limited [ref.6]. 15 The analyses and conclusions of this chapter have been worked in consultation with the EUROCONTROL Agency Legal Service. 16 Paragraph 59: implement a non-punitive safety reporting environment in order to assess compliance with an objective safety target approach. 17 At the fifth Session of the Provisional Council (July 1999), the Provisional Council urged all States to adopt no-penalty safety incident reporting policies where these were not already in place. 18 Notes are included in ICAO Annexes to provide information on a standard or recommended practice, but they do not constitute part of the standard of or the recommended practice In addition, the incident investigation and analysis will always require human intervention to ensure that all complexities are addressed. 21 The real categorisation of the accident will only be known after publication of the final accident report. 22 Source: Air Transport Intelligence. 23 Alitalia DC-9, Nov NTSB Safety Recommendations A-92-8 and See also PRR 4, Chapter 3 [ref.10]. 25 Notified by France to the SRC and by the UK CAA to the FSF in the 1998 Jul-Sept Digest. 26 An indirect cause of the accident which occurred in Paris CDG on 25 th May 2000 [ref.4]. 27 Reported as serious problems by the UK and Sweden. 28 Identified by the SRC [ref.6] PRR 5 - Safety

31 4 Delays 4.1 Introduction Civil aviation delays can be viewed from different perspectives: The passenger s view: focusing on arrival delay [ref.15]; The airline view: focusing on delays measured in relation to published flight schedules; The ATM view: focusing on delays measured in relation to the last filed flight plan This chapter deals first with air transport delays in Section 4.2 (the airline view), which sets ATM performance in context ATM performance in terms of capacity and delays is then considered in sections 4.3 and following, with an emphasis on Air Traffic Flow Management (ATFM) delays. 4.2 Air Transport delays This section concentrates on departure and arrival delays of scheduled airlines, based on information from the Central Office of Delay Analysis (CODA, ref.16) 29. A graphic representation of air transport delay components is given in Annex Departure delay (>15 min) 26% 25% Average Departure delay min 39 min Arrival delay (>15 min) 27% 25% Average arrival delay min 40 min Data source: CFMU & AEA Figure 24: Air Transport punctuality The percentage of flights delayed for more than 15 minutes is widely used as an indicator of air transport punctuality (on-time performance). Figure 24 shows this indicator for arrivals and departures, and the ATFM delay contribution to these, for the years 1997 to The table shows that there has been little change in arrival and departure punctuality from Reactionary delays 32 remained the dominant departure delay cause (37%) in 2001, slightly below 2000 levels (39%) ATFM delays accounted for 33% of departure delays in 2001 (excluding reactionary delays), a reduction from 38% in Both Figure 24 and Figure 25 indicate that approximately one third of air transport delays are attributable to ATC. ATFM delays reduced sharply in the September-December 2001 period, mostly due to a decrease in traffic (see 2.2). Delay - PRR 5-31

32 Figure 25: Air transport departure delay causes (excluding reactionary) Data source: AEA Figure 25 also shows that delays attributed to security measures nearly tripled to 11% after 11 September. 4.3 ATFM delays The following sections concentrate on ATM capacity and delays, in particular whether delay targets adopted by the Provisional Council were met ATM delays can be imposed before departure (ATFM delays), during taxi or in flight. In the absence of precise information on taxi and airborne delays due to ATM (e.g. airborne holding), this section concentrates on ATFM delays ATFM delays are broken down according to their origin, i.e. en-route or airport capacity restrictions. En-route ATFM delays are reviewed in sections 4.4 to 4.7, and airport ATFM delays in section 4.8. More details on ATFM delays can be found in the very comprehensive documentation published by the Central Flow Management Unit (CFMU) [ref.17] Insufficient ATC capacity is the dominant cause for en-route ATFM delays. Sub-optimum use of existing capacity also generates delays, which is addressed in section 4.9 (ATFM Performance) Figure 26 shows a clear relationship between traffic and ATFM delays. Delays grow very fast when traffic grows and capacity remains constant. A shift in the traffic/delay curve to the right corresponds to an increase in capacity. The observed difference between week and week-end delays mainly originates from different traffic patterns and a mismatch between week-end demand and capacity. More details on the assessment of traffic, capacity and delays can be found in Annex PRR 5 - Delay

33 Figure 26: Traffic and en-route ATFM Delays (2000, 2001) Data source: CFMU Target and actual delay (European level) The generally agreed Key Performance Indicator (KPI) for en-route delays is the average en-route ATFM delay in the summer period (May to October inclusive). The Provisional Council agreed for the preparation of Summer 2001 to maintain the Summer 2000 target of 3.5 minutes, on the understanding that this would comprise a target of 2.8 minutes for en-route delays Although traffic growth (+0.2%) was well below planned growth (+5.3%), the en-route ATFM delay target set for Summer 2001 (2.8 minutes) was exceeded (3.1 minutes), as shown in Figure 27. Delays would have been some 35% higher 34 if traffic had grown as planned. Figure 27: Summer ATFM Delays In 2001, en-route ATFM delays remained dominant and airport ATFM delays constant at around 0.8 minutes per flight (25% of ATFM delays). This proportion could become more significant if en-route ATFM delays were reduced. It is also widely variable across Area Control Centres (ACC) as shown in Figure 28. Delay - PRR 5-33

34 Figure 28: En-route and airport ATFM delays Key figures for European traffic and ATFM delays are listed in Figure 29. FULL YEAR SUMMER Year Traffic ATFM delays Average ATFM delay per flight Average daily traffic (flights) T r a f f i c vo l u m e ( k m ) index 36 ATFM delays > 15 min Total ATFM delays (min) Total ATFM delays (index) En-route ATFM delays (min) En-route (min/flight) Total Delay (min/flight) Target Total (min/flight) % 20.9M m none % 27.4M m % 43.3M m none % 31.8M m % 27.6M m Data source: CRCO (km) & CFMU (traffic, delays) Figure 29: Key European traffic and delay data 4.4 En-route capacity and delays (European level) In 2001, 64% of ATFM delays were due to insufficient ATC capacity [ref.17]. Effective capacity 37 is defined as the traffic volume (km) which the ATM system can handle with optimum delay, i.e. one minute per flight 38. The average effective capacity continued to lag behind demand and did not increase by much in , as shown in Figure Traffic volume (km) Effective Capacity Data source: CRCO & PRU Figure 30: Traffic and effective en-route capacity 34 - PRR 5 - Delay

35 4.4.2 The capacity objective of the ATM Strategy is To provide sufficient capacity to accommodate the demand in typical busy hour periods without imposing significant operational, economic or environmental penalties under normal circumstances [ref.7]. This has not been met This capacity lag has high cost implications for users and society as a whole, as indicated in Figure 31. Costs of en-route ATFM delays to the European airline industry are estimated to be between 1100 and 1700 M, and between 2500 to 3600 M if costs to passengers are added 40. Airspace users Delay (Minutes) Unit cost ( /min) Total value ( ) En-route ATFM delays 21 M M Reactionary delays 10 M M Total airspace user costs 31 M M Passengers 31 M M Total en-route ATFM delay costs M Figure 31: En-route ATFM delay costs in 2001 Data Source: CFMU & ITA [ref.18] This continued capacity lag indicates a lack of responsiveness within the ATM system on the need to provide adequate capacity. This is the result of a combination of under-capacity in a few locations and over-capacity in other places (see below). At the moment, there is no strong signal to ANSPs as to the right level of capacity A more detailed monthly analysis is shown in Figure 32, where traffic and effective en-route capacity are both expressed in terms of flights 42. Delays are optimum when traffic and effective capacity are equal Flights Jan 1997 Jul 1997 Jan 1998 Daily traffic Jul 1998 Kosovo Jan 1999 Jul 1999 Month Jan 2000 ARN V3/3 Jul 2000 Jan 2001 Jul 2001 Effective capacity NERC Jan 2002 Data Source: CFMU & ITA [ref.18] Figure 32: Effective en-route capacity and traffic Delay - PRR 5-35

36 4.4.6 Several observations can be made from Figure 32: Effective capacity 43 reached and even exceeded traffic demand in the winter traffic trough, but remained far below the summer traffic peak, which is typically 25% higher. Capacity improvements were achieved towards the end of the summer, too late to have a significant effect on year 2001 delays. This phenomenon was already observed in past PRC reports. Capacity must better match summer traffic demand to meet delay targets. This would entail some capacity surplus in the winter, which can be used to train for peak summer loads. The negative impact of the Kosovo crisis is very visible. The temporary downturn and subsequent improvements in major airspace redesign (ARN V3 Phase 3, 19 April 2001) and system implementation (UK new En-Route Centre (NERC), 26 January 2002) also show up. This clearly shows the high interconnectivity of the European network, where restrictions in one area are sufficient to affect the delay performance of the whole of Europe. It also shows that meeting the optimum delay target will not be possible without effective co-ordination among all ANSPs. On average, the yearly traffic increase from 1997 to 2001 was 5%, and the effective capacity increase was 4%. Advantage must be taken of the current traffic slow down to close the capacity gap and meet the ATM capacity objective. 4.5 Network effects (regional level) European regions were defined in PRR 1 for analysing network effects and the impact on delays Figure 33 shows that 65% of ATFM en-route delays were concentrated in London, Area North, and Area South in 2001 (see definitions in glossary). Figure 33: ATFM En-route delays in European regions 36 - PRR 5 - Delay

37 4.5.3 Figure 34 depicts actual and expected 44 delays in European regions in Area South, the most delay-prone region from 1998 to 2000 (see Annex 3), improved as expected in This positive evolution shows the effects of actions undertaken by ANSPs and EUROCONTROL (CHIEF). Area North was the most delay-prone, even more than expected, indicating the importance of the 5-States programme 45. This is analysed in more detail below. Data source: CFMU Figure 34: ATFM delays in European regions Figure 35 shows actual traffic, actual delays and the expected ATFM en-route delays if capacity had been implemented as committed En-route delay (minutes) Jan- 01 Feb- 01 Mar- 01 Apr- 01 May- 01 Jun- 01 Jul- 01 Aug- 01 Sep- 01 Oct- 01 Nov- 01 Actual Delay Expected Delay ACC Mvt Dec Number of ACC movements Data source: CFMU Figure 35: Traffic, actual and expected ATFM en-route delay (2001) The year 2001 can be split into four periods: 1. ATFM delays were more or less as expected until 19 April. 2. ATFM delays increased well above expected levels for about three months after 19 April. This can be related to the temporary negative impact of the new route network ARN V3/3 on capacity. 3. Between 10 July and 11 September, delays were lower than expected. This could be understood as a positive result of capacity enhancement measures taken earlier. 4. The traffic downturn after 11 September resulted in a large reduction in delays (see traffic/delay relationship in Figure 26) The major event in Area North was the implementation of EAM 04 in Germany. The EAM 04 implementation provided clear benefits in Rhein Upper Airspace Area Control Centre (UAC) from July onwards, but these were offset by a temporary deterioration in Maastricht (see Figure 36), where some ATC sectors have been negatively affected by the new airspace structure in the area. Delay - PRR 5-37

38 En-route delay (minutes) Jan- 01 Rhein Feb- 01 Mar- 01 Apr- 01 May- 01 Actual Delay Jun- 01 Jul- 01 Aug- 01 Sep- 01 Oct- 01 Nov- 01 Expected Delay Dec Jan- 01 Maastricht Feb- 01 Mar- 01 Apr- 01 May- 01 Actual Delay Jun- 01 Figure 36: En-route delays in Rhein and Maastricht UAC Jul- 01 Aug- 01 Sep- 01 Oct- 01 Nov- 01 Expected Delay Dec- 01 Data source: CFMU Delays were further exacerbated by complex interactions between eight civil and military ANSPs in the area, by the EAM04 implementation date (in April, when traffic was growing), by related staffing issues within the Deutsche Flugsicherung GmbH (DFS) and by technical problems in Maastricht UAC. The net result of such a complex situation was a deterioration in delays in Area North (Figure 34) in In general, a major problem in the backbone area of Europe (the arc from London to Maastricht and Marseille) is the large workload generated by network inefficiencies, notably: A lack of ATM flexibility in managing capacity to cope with hub-and-spoke operations and/or sudden changes in traffic flows [ref.19]; The wide variety of ATM civil-military arrangements and procedures [ref.20]; A high load of co-ordination tasks due to the presence of numerous ANSPs in the area (par ); The difficulty in reconciling airspace requirements when they conflict with each other at international level [ref.20]. 4.6 ATFM delays (ANSP level) Figure 37 shows those ANSPs that had the most influence on ATFM delays in Summer The figure presents ATFM delays in 2000 and 2001 and the delays which would have been expected if capacity commitments were met. 4.0 Summer 2000 Summer 2001 Expected 2001 En-route delay Million minutes France (DNA) United Kingdom (NATS) Germany (DFS) Switzerland (Skyguide) Maastricht (EUROCONTROL) Spain (AENA) Italy (ENAV) Data source: CFMU Figure 37: Summer ATFM en-route delay per ANSP 38 - PRR 5 - Delay

39 4.6.2 Some ANSPs, e.g. the French ANSP, were able to estimate capacity increases from ATM improvements with good precision. In several other cases, there remain large differences between committed and actual capacity Italy performed better than expected, mainly due to successful actions in Padova ACC. Delays in ECAC would have been higher if Padova had not provided much more capacity than committed. Capacity gains could probably have been better anticipated The case of Germany and Maastricht is discussed in In the UK, London ACC delays were higher than expected due to the preparation of the transfer to the new ACC (NERC). In both cases, complex ATC capacity implementations entailed temporary capacity restrictions Staffing issues remain a predominant constraint in many parts of Europe. These also have an impact on the European ATM network 46. There is some recruitment of controllers from abroad in order to meet staff shortages. Only well co-ordinated transfers of controllers across ACCs will help solve the ATFM delay situation, e.g. temporary employment of Irish controllers by the German DFS Capacity management has significantly improved over the last three years, but more remains to be done. 4.7 Delay targets and capacity ECAC targets ECAC-level targets are set in terms of delay. The traffic volume risk is therefore borne by the European ATM system. The delay target is set with reference to the generally agreed Key Performance Indicator (KPI), i.e. average en-route ATFM delay in the summer period (May to October inclusive) The EUROCONTROL Provisional Council endorsed an overall ATM network target to reduce progressively to a cost-optimum average ATFM delay of one minute per flight by the Summer 2006, as a basis for the co-operative planning and provision of capacity 47. For 2001, the Summer en-route ATFM delay target had been set at 2.8 minutes Figure 38 depicts summer ATFM delay targets decreasing progressively between the agreed targets for 2001 and These targets should remain independent of traffic variations, so as to meet the ATM Strategy capacity objective (capacity to meet demand under normal circumstances). En-route delay per Flight (minutes) ,9 4,1 5,5 3,6 3,1 Actual : 3.1 min/flight Target : 2.8 min/flight Summer Optimum delay Target delay Actual delay Figure 38: Summer en-route ATFM delay target Data source: CFMU Meeting the delay target In order to meet this target, the capacity gap (see Figure 30) should be closed and future traffic growth be accommodated. This is a challenge. Many elements such as new airspace designs, traffic flow changes, success and timing of major implementations, staffing issues, routing schemes, all play a role in achieving performance targets. Delay - PRR 5-39

40 4.7.5 Experience has shown that delays are very sensitive to the traffic/capacity relationship, and that ATM performance can be compromised by specific issues. This was the case in 1997, 1998 and 1999 with issues in Greece, Area South and Kosovo respectively. In 2001, temporary capacity restrictions linked with the implementation of ARN V3/3 (EAM 04) and NERC transition plans had a substantial impact on delays, but were an investment for the future. This impact is minimised when transitions occur during traffic troughs The magnitude of such events is amplified by the lack of flexibility of the ATM system: (1) rigidity of the ATFM system which accounts for 25% of delays (see 4.9.6); (2) rigidity in the use of airspace by civil and military traffic (see chapter 10); and (3) persistence of critical staffing issues (ATFM delays related to collapsed ATC sectors) The capacity gap must be closed as soon as possible, given the high cost of delays 49. The present traffic slowdown offers a unique opportunity to close the capacity gap, so as to reach and maintain the delay target The cumulated capacity growth in 2001 and 2002 may well be close to the capacity gap, which would result in low delays in However, traffic is forecast to increase again in There must be no reduction in effort to increase capacity to meet and maintain the optimum en-route ATFM delay target (1 minute per flight). Meeting this challenging, yet achievable target, requires well co-ordinated efforts from ANSPs and the EUROCONTROL Agency. ANSP targets EUROCONTROL translates the ECAC delay target into capacity targets for individual ACCs, using the FAP methodology. ANSPs then indicate capacity increase commitments for each of their ACCs, which are recorded in Local Convergence and Implementation Plans (LCIP) Figure 39 shows capacity targets and commitments for 2006 for European regions 50. The numbers in the Figure need to be confirmed, as post-11 September traffic forecasts are uncertain. This is particularly the case for Area South, Area North and Paris, where capacity targets have been lowered significantly The latest capacity plans and traffic forecasts indicate an excess of capacity in the South East area in 2006 (+34%) and a capacity shortfall of 21% in Greece (which may be improved by enhancing civil-military co-ordination). In 2006, terminal areas may become critical in Europe. Transition plans to implement Central European Air Traffic System (CEATS) in 2007 should include plans to reduce excessive ATC capacity, which may result after the transfer to the new ACC PRR 5 - Delay

41 Area 2006 Capacity Target (ACCmov/hr) 2006 Committed Capacity Plan (ACCmov/hr) Difference between committed and target capacity for 2006 Area North % Area South % South-East % Benelux % CEATS % Central Mediterranean % France West % Germany % Greece % Iberia % London % North Europe % Paris % Data source: LCIP 2001 Figure 39: Required and committed level of capacity There has been progress in capacity planning under EUROCONTROL auspices, as discussed above but ANSP capacity plans remain indicative, non-binding and adopted unilaterally. The result is a combination of capacity shortfalls, which generate delays, and of excess capacity, which generates excess costs Whereas capacity cannot be increased at once, it should be possible to design and jointly agree coherent medium-term plans (staffing, airspace design, investments, etc.) which together meet ECAC targets. Delay - PRR 5-41

42 4.8 Airport ATFM delays Issues associated with airport ATFM delays are quite different from en-route. The slot co-ordination process caps the scheduled demand at airports, whereas en-route ATFM delays are used to keep demand within available capacity. As a consequence, ATFM measures are generally applied at airports only in exceptional circumstances (bad weather or technical failure) Figure 40 indicates that airport ATFM delays were imposed mostly to control arrival flows. The most penalising airports for ATFM delays are listed in Annex 2. Figure 40: Airport ATFM delays (2001) ATFM delays are imposed several hours before arrival, which does not allow precise arrival flow control and quick response to unforeseen changes. Very different techniques are applied in the US (rerouting and en-route flow control using miles in trail, while ground holds are used only in exceptional circumstances) Such techniques, however, require co-ordination across centres, which generally does not exist in Europe. At the moment, there is no incentive and no supporting automation for centres (e.g. Geneva, Reims) to pre-sequence aircraft into terminal areas (e.g. London). Upstream centres could be encouraged 51, equipped, and even remunerated, to optimise Terminal Manoeuvring Area (TMA) throughputs and ease their workload The distribution of airport ATFM delay causes, namely ATC delay, aerodrome (non-atc), weather and others, is shown in Figure 41, for years 1997 to Data source: CFMU Figure 41: Airport ATFM delay causes Total airport ATFM delay fell in 2001 (a decrease of 8.6% on the previous year), after increases in the three previous years. Figure 41 shows that ATC capacity is a small part of airport ATFM delay causes, which has decreased further in The proportions of ATFM delay causes have changed over the last five years. With the exception of 1998, aerodrome non ATC delay increased year on year whereas ATC delay decreased. Most significantly, aerodrome delay doubled from 1999 to 2000 (11% to 22%), and ATC delay almost halved from 2000 to 2001 (from 31% to 18%) PRR 5 - Delay

43 4.9 ATFM performance Air Traffic Flow Management (ATFM) seeks to protect ATC sectors and airports against traffic demand in excess of declared capacity. It does so using AFTM measures such as departure delays according to set rules (e.g. first planned, first served) There are two key factors when examining the performance of the ATFM system 52 : 1. The degree of protection against excess demand. This can be measured with the number of over-deliveries The optimum use of available ATC capacity. This can be measured with lost ATFM slots and unnecessary ATFM regulations. Over-deliveries Figure 42: Over-deliveries Figure 42 shows the proportion of cases when traffic exceeds declared capacity (over-deliveries) for initial demand (blue bars) and actual traffic resulting from ATFM processes (yellow bars). Over-deliveries in excess of 10% remain significant (7%), and can at times exceed 30%. Declared capacities are then understated by corresponding amounts so as to preserve safety. If over-deliveries could be better controlled, declared capacities could be raised accordingly A major cause of over-deliveries is non-adherence to ATFM slots (see Figure 43). More specifically, nonadherence tends to concentrate at certain airports and involve specific airlines. The most frequent non-adherence reports for airports and airlines are listed in Figure 44. Source: CFMU Figure 43: Non-adherence to ATFM slots Delay - PRR 5-43 Source: CFMU

44 Airports Non-compliance with ATFM slots Airlines Non-compliance with ATFM slots Budapest/Ferihegy 52% AIR One 32% Brussels 46% Iberia 31% Venice/Tessera 34% Air Canada 30% Madrid/Barajas 33% American Airlines 28% Milan/Linate 32% Alitalia 28% Data source: CFMU Figure 44: Top airports and airlines in non-adherence reports Rules governing ATFM slots, including non-compliance procedures, should be included in the General conditions for the operation of the common European ATFM system 55. The PRC has already proposed a non-compliance escalation process 56 : (1) Non-compliance reports submitted by the CFMU to non-compliant parties, (2) Review of unresolved cases and publication by the PRC, (3) Legal enforcement actions by the airspace regulatory authorities. The PRC proposal should be addressed under the ATFM action plan (action 1). Capacity utilisation According to CFMU statistics, some 25% of ATFM delays are caused by unnecessary regulations (i.e. demand never exceeds capacity) and by unused ATFM slots in congested periods. This costs some M 57 annually to airspace users Most ATFM delays occur with regulations being applied for low levels of excess demand. Ground delays (i.e. ATFM regulations) are not well adapted in such cases: random effects and tolerances 58 result in poor ATFM performance (many unused slots, and yet some over-deliveries). It should be noted that the US ATFM system rarely uses ground delays to solve en-route capacity constraints. Instead, re-routing and airborne flow control are generally applied. Specific flow control tools need to be applied in the many cases where demand slightly exceeds capacity. The situation is already being improved and could possibly be further improved in at least two ways: 1. Improved tactical flow management: in particular through the introduction of ETFMS in early Re-routing: a simulation carried out by EUROCONTROL showed that a substantial proportion of ATFM delays could be suppressed with re-routing or flight level changes. Re-routing and level capping are not used extensively in current ATFM activities (re-routings offered to less than 1% of regulated flights on a tactical basis 60 ) In summary, the ATFM system is not as effective as it could be in protecting ATC sectors and in making use of available ATC capacity. This is largely due to abuses of ATFM slots 61, to ATFM techniques such as re-routing remaining under-used, and to the fact that the CFMU has no clear mandate for tactical flow management at European Level (action 1 of the ATFM action plan) Conclusions Although traffic grew by only +0.2% during the Summer period 2001 (May to October), well below planned growth (+5.3%), en-route ATFM delays (3.1 min) were above the interim target (2.8 min). Had traffic grown as expected, delays would have been even higher (an estimated 35% above) The capacity gain in 2001 was substantial but came too late to have a strong influence on summer delays. Year 2002 is expected to be better, due to relatively low levels of traffic and the positive effects of airspace structure and system implementations PRR 5 - Delay

45 Effective capacity did not increase sufficiently and has lagged behind demand over recent years. This indicates a lack of responsiveness within the European ATM system under the present institutional and incentive arrangements. According to the ATM objective, capacity should meet demand under normal circumstances. The cost of ATC delays to airspace users was in the order of 1100 to 1700 M in ATFM delays were mostly due to insufficient ATC capacity, particularly in Area North and London in the high traffic season. European ATM network inefficiencies and staffing issues appear to be the most critical factors behind these delays There should be no reduction in effort to increase capacity to meet and maintain the optimum enroute ATFM delay target (1 minute per flight). Action plans to this end should be incorporated in relevant ANSPs business plans It is not just a matter of ATC capacity; the ATFM system is not using existing capacity to best effect. Some 25% of ATFM delays are due to unnecessary ATFM regulations and lost slots. In addition, the ATFM system does not fully protect sectors against over-deliveries, leading to understated declared capacities. This is largely due to abuse of ATFM slots, ATFM techniques such as re-routing remaining under-used, and the fact that the CFMU has no clear mandate for tactical flow management at European level. Binding rules of application for ATFM and a review of ATFM principles and operating modes are needed. This must be addressed urgently as part of the ATFM action plan. Delay - PRR 5-45

46 Notes on chapter 4 29 There is presently no information from CODA on non-scheduled flights. 30 Flights delayed >15minutes. 31 Flights delayed >15minutes. 32 Reactionary delays are due to late arrival of aircraft or crew from previous journeys. 33 Decision taken at the ninth session of the Provisional Council (PC 9) in November Considering a traffic/delay elasticity of 7 [ref.17]. 35 In certain cases (e.g. Athens) airport ATFM delays are also generated by capacity constraints in the terminal airspace around airports. 36 For States participating in the EUROCONTROL Routes Charges System in The underlying methodology is presented in Annex See PRR 4, section 4.12 [ref.10]. 39 Estimated capacity increase: +1.6% (CFMU briefing to CESC/5). Most penalising locations are listed in Annex Delay costs are computed using unit costs published in ref Reactionary delays being approximately one third of delay causes (see 4.2.3), are approximately one half of primary delays. 42 The underlying methodology is presented in Annex This corresponds to an average en-route ATFM delay of 1 minute per flight. 44 The underlying methodology is presented in Annex See item 25 in ref See ref.17 (delays due to staffing issues and collapsed ATC sectors). 47 Results of the 10 th session of the Provisional Council held on 5 April Targets of 3.5 minutes per flight for summer ATFM delays, which were set for 1998 and 2000, are equivalent to a target of 2.8 minutes for en-route ATFM delays set at PC 9 (Nov. 2000). 49 See See glossary and Figure The Independent study for the improvement of ATFM (EUROCONTROL, September 2000) has recommended the development of positive indicators, in order to develop a Single Sky spirit and solidarity between all the ATFM actors, have a better balance of ATC workload and aim at a global optimum instead of local ones. 52 ATFM slots are issued some two hours before departure. Inevitable perturbations to the plan and breaches of ATFM rules (see Figure 43) result in bunching (over-deliveries) and empty space (lost slots). 53 An over-delivery occurs when the actual number of aircraft that enters the sector during a particular period exceeds the regulated capacity. An over-delivery does not necessarily result in an overload. An overload occurs when an ATCO reports that he/she has had to handle more traffic than he/she considers to be safe. 54 See Performance Review Report 2 (PRR 2), section Article 7.2 (d) of the EUROCONTROL revised Convention. 56 See PRR 4, [ref.10]. 57 Using same delay costs as in Figure e.g. the tolerance window of 15 minutes for the take-off time (ATFM slot). 59 Enhanced Traffic Flow Management System. 60 ATFM/IFPS operations statistical analyses, EUROCONTROL/CFMU. 61 This is being addressed under action 6 of the ATFM action plan PRR 5 - Delay

47 5 Cost-effectiveness 5.1 Introduction Cost effectiveness is the third main Key Performance Area. The provision of cost effective air navigation services is becoming even more important in light of the current difficulties experienced by the air transport industry. 5.2 Key Performance Indicators (ECAC level) This section reviews en-route costs in 2000, and the forecast for 2001 and 2002 (see Figure 45) P 2002P 00/ /02 63 Contracting States Total en-route costs (M ) % 12% National costs % 12% EUROCONTROL Maastricht % 28% EUROCONTROL Agency % 7% Kilometres (million) % 2% Nominal unit cost ( /km) % 10% Price index % 3.6% Real unit cost 65 ( 2000/km) % 6% Figure 45: Evolution of en-route costs Data source: CRCO Final cost data for 2000 shows a slight decrease in average unit costs for the sixth consecutive year. This virtuous trend is clearly over, as indicated by the projected increase in the unit cost for 2001 and This trend already apparent in last year s projection (see PRR 4) has been accentuated by the traffic downturn observed in the last quarter of Data source: CRCO Figure 46: Evolution of real unit cost, total costs and traffic Cost-effectiveness - PRR 5-47

48 5.3 Unit costs per State Unit costs are obtained by dividing the en-route costs (including EUROCONTROL costs), as declared by the States to the EUROCONTROL Central Route Charges Office (CRCO), by the number of kilometres computed by the CRCO. This is considered as a better measure of ATM performance than the unit rate used for charging purposes. First, the value for a given year is not affected by over- or under-recovery in the previous year. Second, the measure is not influenced by aircraft weight Figure 47 shows the 1998 and 2000 unit costs for each State and the projected 2002 unit costs, based on the information declared by the States for the establishment of the 2002 unit rate (red lines). Data source: CRCO Figure 47: Unit costs per State Evolution from 1998 to Unit costs decreased in 13 States and increased in 12 States between 1998 and The most significant decreases in unit costs were recorded in Ireland, France, Italy, Slovenia, Austria, and the Netherlands. High increases in the cost per km occurred in Greece (+50%), Croatia (+44%), Portugal (+27%), and Hungary (+25%). Projections for Projections for 2002 show a decrease in unit costs in only 10 States; the most significant reductions are expected in Bulgaria, Turkey and Austria. Unit cost increases are foreseen in most States; most significantly in Switzerland, Hungary and Belgium-Luxembourg As shown in Figure 48, unit rates are planned to increase in all but five States. In November 2001, the enlarged Commission decided to freeze the unit rates at their 2001 level for the first quarter of Unit rates will however be adjusted in the second part of the year to compensate for revenue shortfalls during the first quarter, thus resulting in the year-on-year variation of the unit rate ranging from -32% to +42% PRR 5 - Cost-effectiveness

49 States * States * Portugal Santa Maria % Sweden % Portugal Lisboa % Denmark % Belg.-Luxembourg % Greece % Cyprus % Italy % Spain Continental % Czech Rep % Hungary % FYROM % Spain Canaria % Moldova % France % U.K % Germany % Slovak Rep % Switzerland % Slovenia % Netherlands % Bulgaria % Ireland % Croatia % Norway % Malta % Romania % Turkey % Austria % Finland Weighted average (Finland excluded) % * Basic unit rates valid for the last three quarters of Figure 48: Change in unit rates between 2001 and 2002 Data source: CRCO Increases in the unit rate are the results of the traffic downturn and/or the lack of cost containment measures. To compensate for the expected loss of revenues, most of the ANSPs/States have increased their unit rate Between June 2001 when first cost estimates were published and November 2001 when unit rates were approved, overall traffic forecasts have been revised downwards by 5%. Costs chargeable to airspace users have decreased by only 1%, resulting in an average increase in the unit rate of about 4% The adjustment mechanism and the carry forward of over or under-recovery experienced in 2000 is adding additional perturbation to the system and contributes in amplifying unit rate variations To justify the increase in the unit rate, it is generally argued that (1) most ANSP costs are fixed in the short run; (2) that the current charging principles do not give the ANSP the flexibility to build up reserves in years of strong traffic growth; and (3) that access to borrowing is either constrained or will ultimately result in additional financial costs for the users On the other hand, it is important to note that (1) a significant part of an airline costs are also fixed in the short run; (2) airlines are not in a position to increase their fares in the current market conditions; and (3) that because of the business risk, the cost of borrowing is more expensive for an airline than for an ANSP The full cost recovery regime provides very few incentives for ANSPs to reduce their costs. Given the significant changes that have occurred in recent years, in particular the corporatisation process and the fact that ANS provision is increasingly considered as a normal commercial service, the question of whether the full cost recovery regime remains appropriate should be re-examined. Cost-effectiveness - PRR 5-49

50 5.4 Gross margin Figure 49: Cost breakdown by type (2000 Data) The gross margin (or operating profit) measures the difference between total revenue and total operating expenses, i.e. staff costs and operating costs according to submissions made by the participating States to the Route Charges System. Under the full cost recovery mechanism, the gross margin is fully consumed by two main categories of costs: depreciation and interest. Interest is a combination of actual interest paid and a fair return on own assets In 2000, the gross margin continued to decrease slightly to 20.7%. This is nearly six points below the 1996 level, most of which is due to a decrease in interest There remain huge variations across States as can be seen in Figure 50. The operating expenses represent nearly 90% of the cost base in Sweden and Ireland, and less than 60% in Turkey and the Czech Republic. Data source: CRCO Figure 50: National costs breakdown by type (2000 data) 5.5 EUROCONTROL costs Agency costs (excluding Maastricht and CRCO) EUROCONTROL Agency costs, 316 M in 2000, represented some 7% of the total en-route costs. These costs are included in the cost base of the 30 Member States and therefore in the Route Charges paid by the airspace users to the Member States. Member States are responsible for approval of the EUROCONTROL budget and for paying their contributions PRR 5 - Cost-effectiveness

51 million Euro 2000 (deflated) % % change over % 23% 20% previous year % 10% 10% 10% 1% 4% 3% 0% Euro / km Figure 51: Evolution of EUROCONTROL Agency costs (Part I) Data source CRCO Figure 51 shows the evolution of the Agency costs. The significant cost increases during the years coincided with the launch of the European Air Traffic Control Harmonisation and Integration Programme (now The European ATM Programme, EATMP). EUROCONTROL membership more than doubled during the 1990s from 13 Members States in 1990, to 30 Member States in Over recent years, the EUROCONTROL cost base has been approximately in line with the overall costs of the European Route Charges System (i.e. around 7%). The right hand side of Figure 51 suggests that the unit cost indicator is projected to level off at around 0.05 per km The breakdown of the Part I cost base per establishment is shown below: Data source CRCO Figure 52: Breakdown of Agency costs per establishment 68 EUROCONTROL Maastricht costs The unit cost for the Maastricht UAC was 0.2 /km in This is significantly below the European average (0.71 /km). It should be noted however that the two figures are not strictly comparable as the Maastricht cost base does not include the costs of meteorological services, neither does it include the costs of the Surveillance and Navigation infrastructure, which are borne by the delegating States Until 2000, Maastricht costs have increased less rapidly than traffic thus resulting in a steady decrease in the unit cost. Forecasts for 2001 and 2002 are far less favourable mainly due to investments for the new control room to be operational in 2002 and to an increase in staff numbers. Cost-effectiveness - PRR 5-51

52 million Euro 2000 (deflated) -5% -1% 3% 4% 7% 5% % change over previous year 11% 11% Data source: CRCO Euro / km Figure 53: Evolution of EUROCONTROL Maastricht costs Data source: CRCO CRCO costs CRCO costs / Amount billed 0.7% 0.6% 0.5% 0.4% 0.3% 0.2% 0.1% 0.0% Data source CRCO Figure 54: Evolution of CRCO costs The CRCO s actual net costs (route charges collection costs) were 19.1 M in 2001, which represent 0.43% of the total amounts billed. 5.6 Actual 2000 costs compared to earlier plans Figure 55 compares actual 2000 with plans made in According to Route Charges principles, States should provide five-year projections of their planned costs. In 1997, nine States failed to comply with this basic information disclosure requirement. Figure 55 suggests that States which did not provide forward-looking information in 1997 tended to experience the largest cost increases For those States which provided the relevant information, the planned cost increases were generally below the planned traffic increases, thus resulting in a reduction in the planned unit costs. Netherlands (- 10%), Malta (-7%), Italy (-6%) and the UK (-4%) were the four States with the most significant reduction in planned unit costs. It is commendable that in all four cases, their objectives have been achieved and in some cases exceeded Actual variations in costs over the period have generally been in line with plans. Actual cost increases are below the plan in four cases and within 5% of the planned costs in six. The most significant exceptions are Portugal (planned costs increase +9%; actual costs increase +58%), Spain Continental (planned +11%, actual +32%), Spain Canarias (planned +10%, actual +22%), Maastricht (planned +13%, actual +22%) and Netherlands (planned +9%, actual +19%) Reliable traffic predictions appear to be more difficult. Over the three-year period, actual traffic increases have been significantly higher than most forecasts. This is probably because financial previsions tend to be based on prudent forecasting. However, traffic increases have been high throughout this period. Higher than expected traffic growth has contributed to more delays and facilitated the decrease in unit costs PRR 5 - Cost-effectiveness

53 Country Planned variation Actual variation Traffic 1 Cost 2 Unit Cost Traffic Cost Unit Cost Greece 11% 15% 95% 70% Slovenia 190% 76% -40% Hungary 2% 51% 48% -3% 61% 67% Portugal 9% 33% 58% 19% Slovak Rep. 69% 54% -9% Czech Rep. 94% 52% -22% Cyprus 17% 24% 37% 11% Spain 11% 27% 32% 4% Maastricht 13% 29% 22% -5% Canarias 10% 18% 22% 4% Italy 28% 20% -6% 33% 22% -9% Denmark 17% 15% -1% 23% 20% -2% Austria 13% 30% 15% -5% 20% 27% Norway 21% 19% -1% Netherlands 22% 9% -10% 36% 19% -13% Germany 16% 17% 0% 20% 17% -2% France 13% 15% 2% 25% 13% -10% Sweden 10% 9% -1% 19% 12% -6% Malta 21% 12% -7% 33% 12% -16% United Kingdom 14% 10% -4% 25% 11% -11% Ireland 13% 13% -1% 28% 8% -16% Switzerland 15% 8% -7% Belgium-Lux. 10% 17% 6% -9% Turkey 3 3% -15% -17% 1 Variation in the number of service units between 97 and 2000 according to plan made in Variation in local currency of the National costs between 2000 and 1997 according to plan made in Cost variation for Turkey is given in Figure 55: planned costs compared to actual Source: CRCO Changes in traffic flows are also at the origin of major changes in traffic trends recorded in certain States. While a 13% increase in traffic was planned in Austria, the actual 2000 figure was 5% below the 1997 level. This compares to a 69% traffic increase in the Slovak Republic and a 94% traffic increase in the Czech Republic over the same period. Cost-effectiveness - PRR 5-53

54 5.7 Benchmarking of European ANSPs Figure 47 illustrates significant differences in unit cost across States. Part of this difference can be explained by the complexity of the traffic. ATC services are obviously more demanding around main hubs where traffic density is high and where most flights are climbing or descending. It is therefore important to appreciate whether the observed unit cost differences are justified by using objective measures of traffic characteristics The benchmarking model introduced in PRR 3 compares the en-route costs to the main traffic characteristics. But other factors not included in the model may influence costs: (1) differences exist in the economic environment in which ANSPs operate: exchange rate fluctuations and cost of living are factors that can have an influence on costs; (2) differences exist in the allocation of costs between en-route and terminal ANS: apparently high en-route costs may in some cases simply indicate that the share of total ANS costs passed to en-route is above the average; (3) the quality of service is not captured in the benchmarking model The model proves, however, to be valuable and robust in assessing to what extent the costs charged by States are related to traffic characteristics. The model has been calibrated on 19 States 69 over the period and takes into account work performed by the PRU and the EUROCONTROL Experimental Centre (EEC) on traffic complexity Density can be defined as the ratio between the total CRCO distance (km) and the surface of the charging area (km 2 ). An average density value might however be meaningless if not evenly spread over the area, and therefore the density indicators used in the model take into account the spatial concentration of traffic within the area The added complexity of handling climbing and descending flights was captured in previous versions of the model using the percentage of over-flight. The numbers of flight levels crossed per kilometre and airport movements per kilometre appear to better represent vertical evolutions. It should be noted that the two indicators are closely correlated, as each airport movement is associated with a climbing or descending phase This model has been simplified in comparison with the model used in PRR 3 and PRR 4 by dropping the average route length, which is correlated with the surface and therefore density of traffic Figure 56 shows unit costs estimated by the model and actual unit costs for the year Euro / km United Kingdom Germany Belgium-Lux. Switzerland France Netherlands Italy Austria Actual Spain Sweden Denmark Hungary Turkey Estimated Czech Rep. Greece Norway Romania Bulgaria Ireland Portugal Cyprus Figure 56: Actual versus estimated unit cost per km (2000 data) 54 - PRR 5 - Cost-effectiveness

55 5.7.8 Unit costs of six States are significantly higher than estimated (see Figure 56): Bulgaria: The unit cost of Bulgaria (the second highest unit cost in 2000) seems to bear little relation to its traffic characteristics. Portugal: The unit cost of Portugal increased significantly in It should be noted that, as indicated in section 5.6, Portuguese 2000 costs are nearly 50% above Portugal s own 1997 forecasts. Romania: The actual unit cost is 44% above the estimated value. This is a significant change from 1999, when exceptional traffic conditions due to Kosovo contributed in reducing the unit cost. Austria: The actual unit cost is 35% above the estimated value. This is however a net improvement compared to Changes in the route network had a negative impact on traffic in Austria. Figure 55 shows that traffic in 2000 is some 20% below the forecast made in Strict cost control appears however to be effectively in place. Austria is the only State where total costs have decreased in real terms compared to 1998, and a further reduction is planned for United Kingdom (UK): The estimated unit cost is highest for the UK, which is understandable in view of its traffic characteristics. Actual UK unit costs are however 31% above the estimated value, with a sizeable increase compared to This change is largely due to monetary factors (8% appreciation in value of the pound sterling). Turkey: The actual unit cost is 14% higher than the estimated unit cost. Compared to the difference noted in 1999 (+41%), this is a substantial decrease States with the lowest unit costs in relation to estimated costs are Hungary (-22%), Netherlands (-19%), Ireland (-19%), France (-15%) and Germany (-14%). On this basis, a reduction of some 15-20% in the European average unit cost would appear to be feasible. Some M could be saved every year if the cost-effectiveness of all ANSPs was at best European levels. 5.8 Information disclosure Since its creation, the PRC has advocated the need to improve the quality and the quantity of information that ANSPs should disclose. The PRC, in close co-operation with a number of ANSPs, has therefore developed an agreed Specification to support the analysis of performance in the area of cost effectiveness and productivity. It covers three elements comprising Inputs (staff and capital), Services produced, and Services used (see below) Subsequently, in November 2001 the Provisional Council adopted the PRC Specification for Information Disclosure. From 2002 onwards, the provision of data will become mandatory, with the first submission relating to the year 2001 due by 15 July Publication of annual reports and business plans should be seen as an essential element of information disclosure. Most, but not all, European ANSPs publish annual reports, and the quality and quantity of information they publish varies greatly. Perhaps even more importantly, very few ANSPs publish comprehensive forward looking plans. Cost-effectiveness - PRR 5-55

56 5.9 Full cost recovery principle September events once again indicated that the air transport industry is not free of business risks. The analysis in section 5.3 shows that risks are unequally shared between the different players. While many airlines have had to take drastic measures to reduce their costs and adapt to the sudden reduced demand, most European ANSPs have increased their unit rates to preserve their en-route revenues The inadequacy of the full cost recovery principle to provide an effective answer to a crisis has led many stakeholders to challenge the system. It is also the PRC view that: The full cost recovery system provides little incentive for ANSPs to control their costs. Any difference between the planned costs and the actual costs (cost savings or overspending) are transferred to the users. The year-end profit of an ANSP does not therefore depend on its operational performance, but only on the difference between the cost of capital charged to the users and its actual cost of capital. The full cost recovery system also provides little incentive for ANSPs to adjust to various levels of demand. The demand risk is mostly borne by the airspace users who experience longer and more numerous delays when traffic demand is higher than expected, or higher en-route charges when traffic demand is lower than expected Pricing was adopted by the EUROCONTROL Commission as one alternative option under route charges principles, provided this option is operated under a regulated environment. Price capping, together with ANSP autonomy from the public sector budgeting rules, gives ANSPs strong incentives to be cost-effective and responsive to airspace users needs The Principles and application of alternative charging mechanisms should be investigated with urgency Cost target The analysis presented in section 5.2 shows that unit costs have decreased by approximately 10% between 1995 and Unit costs are planned to increase by 6% in real terms between 2000 and 2002 (see Figure 45) While much needs to be done to progress our understanding of performance, initial analysis shows that performance varies widely across Europe and that scope for improvement exists in at least two areas: Cost-effectiveness of individual ANSPs. Some M could be saved every year if cost-effectiveness of all ANSPs was at best European levels (see 5.7.9). Cost-effectiveness of the European ATM system as a whole. PRR 4 showed that the unit cost of the European ATM system (en-route and terminal) is some 70% above its US counterpart. Part of this difference may be due to fragmentation of airspace, civil and military ANSPs, ATS units and systems, which is discussed in chapter 9, but not yet quantified Even if only a part of potential savings can realistically be achieved, the cost-effectiveness issue appears to be of the same order of magnitude as the delay issue (one billion Euro range). While this conclusion must be treated with great caution, it indicates potential considerable gains. Such gains cannot be achieved at once, but a progressive reduction in the average unit costs is desirable and achievable if effective decisions are taken on Single European Sky issues, including fragmentation While a performance target has been defined for delays, no target exists yet in the area of cost effectiveness. Figure 57 gives a qualitative illustration of the present unit cost trend and desirable trend PRR 5 - Cost-effectiveness

57 Figure 57: Unit cost trend (qualitative) Significant improvement in the target setting process needs to be introduced both at the European and national levels: A high level target in terms of average unit cost should be defined and approved by the Provisional Council. The PRC will submit proposals to this effect. Individual ANSP performance objectives should be clearly spelled out (forward looking part of information disclosure), consolidated at European level and compared to the performance targets set by the Provisional Council, so that network issues can be identified and addressed (see chapter 11). Actual performance should be continuously monitored and compared to the plan Conclusions Lower traffic growth has had a negative impact on unit costs. In this context, ensuring cost-effective ATM is of the utmost importance The recent upward trend in en-route unit costs is raising concerns. The current cost recovery system does not provide incentives for ANSPs to deliver the right level of capacity and to control their costs. It is also unable to cope adequately with a sudden change in traffic. The Principles and application of alternative charging mechanisms should be investigated with urgency, together with other measures to improve ANSP flexibility and responsiveness To ensure a balanced approach, it is important that the Provisional Council sets European-wide objectives for delays and unit costs. It is even more important that each ANSP defines and is committed to clear performance targets. A significant improvement in the target setting process needs to be introduced both at European and national levels. The PRC will propose to the Provisional Council cost performance targets for the European ATM system PRR 4 showed that the unit cost of the European ATM system (en-route and terminal) is some 70% above its US counterpart. From this and benchmarking within Europe, the cost-effectiveness issue appears to be of the same order of magnitude as the delay issue. While this conclusion must be treated with great caution, it indicates potential considerable gains if effective decisions are taken on Single European Sky issues The PRC has drawn up, and the States have agreed, a Specification for Information Disclosure. This is enabling benchmarking, an essential tool for ANSP management, ATM regulation and performance review. The experience gained is shared world-wide with interested States/ANSPs, with the aim to achieving maximum commonality of reporting and performance analysis. Cost-effectiveness - PRR 5-57

58 Notes on chapter 5 62 Evolution computed for the 27 States participating in the Route Charges System in Evolution computed for the 28 States participating in the Route Charges System in EUROSTAT MUIC: Monetary Union Index of Consumer Price (January 2002 index). 65 Real unit cost is deflated from the nominal unit cost using the price index. 66 The variation is even +85% for Santa Maria but Santa Maria is outside the scope of this report. 67 Unit rates expressed in Euro may vary depending on exchange rates. 68 Filled posts at 1/3/ All States having traffic higher than km in 2000, with the exception of Portugal and Bulgaria. 70 To compute the concentration index, the charging area is divided in a number of cells of equal size. Following an analysis of how the size of the cell impacts the concentration index, a cell size varying between 40 NM and 50 NM was used PRR 5 - Cost-effectiveness

59 6 Flight efficiency 6.1 Introduction Flight efficiency has been identified as one of the Key Performance Areas by the PRC [ref.3]. It influences flight time and fuel burn, and therefore the users costs. It also has an influence on the environment (fuel burn, emissions), which could not be precisely assessed with resources available for this report This chapter focuses on the contribution of ATM to en-route flight efficiency in terms of horizontal route extension. Subsequent PRRs will expand on this and also include work on the vertical flight profiles. 6.2 Flight efficiency: initial performance indicator An initial flight efficiency indicator is the average flight inefficiency. Horizontal flight inefficiency (%) for a particular flight is defined as the difference between actual and optimum flight path (unconstrained by ATM routes) divided by the optimum flight path length Flight efficiency was computed by the EEC for a sample of recorded flights covering approximately Areas North, South and London (see Figure 58). Source: EEC Figure 58: An example of ARTAS recorded tracks Optimum routes were compared with actual flight tracks 72, which yielded horizontal flight inefficiency (see Figure 59) and additional fuel burn (see section 6.3 below). Flight efficiency - PRR 5-59

60 14 12 Horizontal route extension% Figure 59: Flight length and horizontal flight inefficiency distribution As expected, the shorter the flight, the greater the impact of arrival/departure routings. Average flight inefficiency for the core area of Europe varies between 10-12% for very short flights (under 400km) and 8% for routes between 400km and 1000km This is an initial evaluation to quantify flight efficiency, based on a limited data set. However, the data seem to be fairly representative of the main European traffic flows and will be improved as more surveillance data become available through the extension of Air Traffic Management Surveillance Tracker and Server System (ARTAS) and the implementation of the ETFMS More analysis is needed to fully understand the extent of the problem. Achieving very high flight efficiency (i.e. very low route extensions) might mean a trade-off with capacity. Again, further analysis is needed to assess the extent of this issue Airspace design has an influence on flight efficiency. This is illustrated by an analysis of selected city-pairs, a subject that will be further developed in the future Figure 60 shows actual flights between Northern Italy (mainly Milan, Venice and Verona) and Benelux (mainly Brussels and Amsterdam), following routes designed to avoid temporary segregated areas when they are activated. Horizontal flight inefficiency can be quite high, reaching up to 17%. It is likely that better civil-military co-ordination could ease this situation (see also chapter 10) Direct Distance km > PRR 5 - Flight efficiency

61 Source: EEC Figure 60: Flights between Northern Italy and Benelux 6.3 Fuel efficiency In trying to assess fuel efficiency, several factors have to be taken into account before drawing definitive conclusions. From a strict ATM perspective, additional fuel per flight or per kilometre seems to be an adequate performance indicator. However, from the more general perspective of air transport, the question of total system efficiency arises: e.g. is a delayed flight cheaper than one flying non-optimal paths and/or profiles? Attributing climate change influences to fuel burn is fraught with difficulty, as, for example, time of day and altitude are two of the many factors that can affect the impact of identical emissions; There are inevitable overlaps among relevant factors; therefore allocating respective influences is not straightforward The key test as to whether an environmental measure is viable is its total cost to society. An environmental measure may produce a cost for airlines, but may result in substantial savings to society. Flight efficiency - PRR 5-61

62 6.3.3 From information available so far, it appears that horizontal flight path extension induces between 2% and 9% additional fuel burn for the en-route portion of flights. This translates into direct additional costs to users and into additional emissions. Terminal phases of the flight can add to this figure (holding patterns, non-optimal climb/descent profiles, extended low-altitude flying) ATFM measures a priori have a positive influence on fuel efficiency. With pre-departure flow control, aircraft are kept on the ground with engines off until they can fly unhampered to their destination. This is apparently more environmentally friendly than holding or vectoring airborne aircraft. 6.4 Conclusions Initial data indicates that average horizontal flight inefficiency for the core area of Europe varies between 10-12% for very short flights (under 400km) and 8% for routes between 400km and 1000km. Flight inefficiency has an influence on the environment (fuel burn, emissions), which remains to be assessed The PRC will further extend the analyses relative to flight efficiency and the impact on the environment, as more data become available in the near future. Notes on chapter 6 71 The optimum 2D flight path is considered to be the great circle between the end points of the flight. This is a good approximation for short/medium haul flights (average flight length about 800km). This is not true for long haul flights due to wind (i.e. optimum path longer, but faster). 72 Actual flight tracks were cut at the points where the aircraft passes 5000ft, due to constraints imposed by an uneven radar coverage PRR 5 - Flight efficiency

63 7 Airports 7.1 Introduction Airports are an essential part of the aviation network. A gate-to-gate approach to ATM is clearly part of the EUROCONTROL revised Convention and of the ATM Strategy [ref.7]. If and when en-route capacity matches demand, in accordance with the ATM Strategy capacity objective, airport issues are then bound to become dominant. This is already the case in the US Initial information on airports was given in PRR 4 (chapter 7). Additional information given here is limited due to higher priorities having been given to other items by both the EUROCONTROL Agency and the PRU under limited resources The effort devoted to ATM items for airports has so far not been commensurate with the issue at hand, nor with the gate-to-gate strategy For its part, the PRC is planning to conduct an airport survey with special emphasis on airport capacity and its utilisation. It will also review terminal charges (approximately 22% of total ATM costs). This survey had to be postponed to 2003 due to budgetary reductions While the common approach taken in ATM is flight-centred, airports tend to take a ground centric approach (see Figure 61), related more to the airframe than to the flight. This is different from the traditional view taken by ATM, where the entity is a flight. However, it should be understood that the airport itself is not involved in the whole flight, but is linking flight origins and destinations. Thus, airports work to ensure a smooth transition of each aircraft through the airport. Figure 61: Airport operations in the gate-to-gate ATM 7.2 Airport Key Performance Areas Airport and terminal airspace safety issues, in particular runway incursions, are addressed in Chapter 3 (Safety) Information on airport delays is given in sections 4.2 (Air Transport delays) and 4.8 (Airport ATFM delays). Non-adherence to ATFM slots at airports is addressed in section 4.9. As en-route ATFM delays tend to decrease, the relative proportion of airport ATFM delays increases (see Figure 27) From initial information, terminal charges for EUROCONTROL Member States can be estimated at 1400 M, i.e. some 22% of total ATM costs (see Figure 1). More information on terminal charges is expected from Information disclosure (see section 5.8). Airports - PRR 5-63

64 7.2.4 Environment at and around airports is not addressed in this report due to lack of data. Noise and emissions are two areas where ATM could bring positive contributions, by improving ground movements, departure and arrival routes, and reducing unnecessary queuing and waiting with engines running. This subject will be further developed in the future. 7.3 Airport capacity Airport co-ordination is the principal means of balancing capacity and demand at European airports. The number of slot-controlled airports has increased constantly over the last ten years Airport slots interact with airline schedules and ATFM slots. Figure 62 shows a typical example. Initial demand for airport slots from aircraft operators is first reduced through the airport co-ordination process, taking declared capacity into account. Airport slots allocated as a result of this process generally do not exceed declared capacity. Airport delays should therefore only appear in exceptional circumstances. However, the actual demand often exceeds both allocated slots and declared capacity. In such situations, ATFM regulations have to be applied, which result in airport ATFM delays INITIAL DEMAND ALLOCATED Actual 80 Declared capacity _01 01_02 02_03 03_04 04_05 05_06 06_07 07_08 08_09 09_10 10_11 11_12 12_13 13_14 14_15 15_16 16_17 17_18 18_19 19_20 20_21 21_22 22_23 23_24 Figure 62: Example of initial, allocated and actual demand Several European airports show differences between the co-ordinated schedule, the airline timetables and real flights (Such cases were observed at some airports in France, Germany, Greece and Spain, amongst others), which need to be further investigated. The meaning of airport capacity is also understood differently by different stakeholders This reinforces the need for better co-ordination between airport scheduling, airport declared capacity and ATFM processes (see former PRC recommendation in ref.2, item 20). EUROCONTROL is now planning to address this issue and the PRC will monitor progress. 7.4 Conclusions The traffic downturn offers an opportunity for en-route capacity to match demand earlier than foreseen. If this is achieved, airport issues will become dominant, as is the case in the US. The effort devoted to airport ATM-related issues has so far not been commensurate with the issue at hand, nor with the gate-to-gate strategy For its part, the PRC is planning to conduct an airport survey with special emphasis on airport capacity and its utilisation. It will also review terminal charges (approximately 22% of total ATM costs). This survey had to be postponed to 2003 due to budgetary reductions PRR 5 - Airports

65 8 Investment in ATM 8.1 Introduction So far, the PRC has given priority to the more pressing ATM issues, in particular delays (PRR 2). This chapter starts to identify issues related to investments in ATM systems and Research & Development (R&D). Investments in Communications, Surveillance, Navigation (CNS) and infrastructure are mentioned for completeness. Corresponding maintenance costs are not addressed here The chapter also reviews progress on the Common systems and projects, part of the ECAC Institutional Strategy adopted in 1997 [ref.1, appendix 8]. 8.2 Investment in ATM The first issue is the magnitude of investment in ATM/CNS in Europe. At the moment, there is a near total lack of transparency on this subject. Evidence of the amount of investments in ATM/CNS can be found in the cost breakdowns provided by Member States in their annual route charges submissions. As indicated in chapter five (Figure 49), depreciation and capital charges 74 are approximately 20% of ATM costs, i.e. some 1200 M per annum In the absence of any precise information, a rough estimate of investment breakdown is presented in Figure 63. According to industry estimates, some 400 M on average is invested in ATM systems every year. Investment in ATM R&D is in the order of 200 M according to the ARDEP report [ref.21]. The remainder is mostly devoted to CNS and infrastructure (buildings, etc). Total ATM systems R&D Other (CNS, infrastructure) 1200 M 400 M 200 M M Figure 63: Rough European ATM investment breakdown Investment in ATM systems and R&D is therefore no more than 10% of ATM costs, which is low by industry standards. 8.3 Investment in R&D As shown in Figure 64, European ATM-related R&D expenditure was 211 M in 2000, of which more than two-thirds was funded from route charges by EUROCONTROL and ANSPs, and 35 M (17%) from public EC funds [ref.21]. This is a very small portion of investment in aeronautics. Figure 64: European ATM R&D funding Source:ARDEP Investment - PRR 5-65

66 8.3.2 Although some results are promising, so far no major technological breakthrough has reached implementation status 77. Operational procedures and system functionality have not evolved greatly since the 1970s. This lack of effective results raises questions about the size and effectiveness of R&D spending in the ATM context 78. European ATM R&D is presently highly scattered, with an average project size of 0.7M, and a multiplicity of sponsors [ref.21]. Although there is a clear need for ATM R&D, it needs to be organised much more efficiently. 8.4 Implementation of individual ATM systems Several major ATM systems have been commissioned in the last two years - Marseille (France), Frankfurt (Germany), Lisbon (Portugal), Rome, Padua (Italy), NERC (UK) to name a few. This is a welcome step forward, which should provide growth capability for the future. However, nearly all ATM system projects conducted by individual ANSPs in the last ten years have suffered from major problems There have been major delays (up to five years) 79, cost overruns and reduced functionality in most individual ATC systems projects. This is documented in particular in the Convergence and Implementation Plan (CIP) Lessons Learnt Seminar [ref.22]. This is a major issue, which has caused not only high financial penalties 80, but also delays in delivering capacity, which in turn have resulted in ATFM delays These major deficiencies can be traced to several common features of individual projects: All ATM systems are tailor-made, with a small part of components being reused. Even the most recent systems have very little technical commonality. This high level of technical fragmentation increases procurement costs and risks for both manufacturers and ANSPs and results in a wide spread of functional levels. This makes it very difficult to synchronise European implementation plans (e.g. data-link). CIPs assembled by EUROCONTROL are only indicative and non-binding. Tailor-made systems and associated fragmentation are bound to result in large penalties to both airspace users and ANSPs. Timing of programme implementations is not always appropriate. While NERC was introduced in a low traffic period, with a minimum possible temporary negative impact, this is not always the case (see 4.5.6). Many projects have followed a big-bang approach, with systems then remaining unchanged for long periods. The evolutionary approach appears to have been more successful (e.g. France, Spain [ref.23]) The aircraft manufacturing industry was in a similar situation in the 1960s. It has evolved considerably since then, with a few competing manufacturers offering fully developed and standardised products, able to satisfy a wide range of users. Such an approach is now starting to be applied successfully to ATM systems (e.g. the new Australian system TAAATS was on time and on budget). 8.5 Cost of individual ATM systems There are implementation and cost issues with individual ATM systems: Financial transparency of ATM investment is poor. Little financial information is publicly available from ANSP sources and none from CIPs. The PRC can therefore present only limited information regarding the cost of recent investments in ATM systems. Some ATM systems have excess capacity, which generates excessive costs (e.g. South Eastern Europe, see Figure 39). The cost of some ACCs (e.g. UK NERC ~ 1100 M 81 ) appears to be proportionally high compared with others (e.g. German Langen ACC ~ 200 M ) The cost recovery principle gives no indication to ANSPs of the right level of investment, because costs are charged to users regardless of investment effectiveness PRR 5 - Investment

67 8.6 European projects European projects and systems can be broken down into central systems implemented in one location (e.g. CFMU systems, EAD) and common systems made available for implementation by ANSPs (e.g. ARTAS, EFDP, POEMS). This paragraph addresses only the latter ARTAS is a common Radar Data Processing System developed by EUROCONTROL. ARTAS incorporates advanced functionality (radar information sharing across borders, etc), which individual projects would probably not have implemented. As EUROCONTROL owns foreground intellectual property rights, ARTAS is made available to Members States at marginal cost and to other States under licence. The total cost of ARTAS development was approximately 30 M, which is cost-effective if some ANSPs use it. By the end of 2001, 11 ANSPs had committed to using it in operation. It was being tested in 40 sites (see Figure 65), including the FAA. ARTAS Implementation Operational Operational in Pilot Sites (actual and planned) SE FI NO EE RU DK LT LV IE GB NL PL BY FR BE DE LU CH CZ SK AT HU SI CR RO MD UA PT ES IT BA YU AL MK GR BG TR MA DZ TN MT CY SY Upper and Lower Airspace UIR partly covered by MAS/UAC Figure 65: ARTAS implementation map Flight Data Processing Systems (FDPS) are another major building block of ATM systems. In the 1990s, the intermediate and advanced levels of FDPS, which were part of the CIP, did not exist on the market. EUROCONTROL undertook to develop two such systems, to ensure availability of a dual source for ANSPs and to control costs through competition among manufacturers. This complex project was eventually stopped after specifications were completed. Development of competing FDPS was then undertaken as joint projects between partner ANSPs (see section 8.7). Investment - PRR 5-67

68 8.6.4 The Pre-Operational European Mode S Enhanced Surveillance (POEMS) programme involves development of dual industry sources of Mode S stations. Joint financing and control of the project by EURO- CONTROL and a limited number of participating ANSPs seem to have ensured success of the project In all three cases, industry would not develop the required products on its own. European projects should probably be undertaken only in cases where there is a clear cost-benefit case for one common system (see section 8.7). Since ATM systems represent only a small market, it may be that European or inter-continental common programmes will continue to be needed for major undertakings. In this case, use of public funds (e.g. Trans-European networks) should be considered. 8.7 Joint projects There is a recent trend towards joint procurement (e.g. Germany-Switzerland-The Netherlands for Mode S; Germany-Spain and France-Italy for Flight Data Processing). This is a welcome start of a possible virtuous circle: with joint procurement, more standard and better-developed industry products become available. This, in turn, will make standard products better suited, and less expensive for other ANSPs, which increases the demand and willingness of industry to invest in standard products Interoperability however becomes a major issue. There is an essential role for ATM systems interoperability, industry standards, and firm implementation deadlines, as in the avionics industry. Overall systems architecture and a Master Plan need to be agreed with ANSPs, airspace users and manufacturers, under the leadership of a central European body. 8.8 Minimum functional levels and interoperability In many cases, functional evolutions must become available over a large area to be effective (e.g. RVSM, data-link). Synchronisation of functional evolutions (across States and across aircraft) will probably be increasingly critical for meeting ATM performance objectives. Fast hardware and software obsolescence of new technology, which implies more frequent system upgrades, is an opportunity to better synchronise functional evolutions Convergence and Implementation Plans (CIP) are published by EUROCONTROL. CIPs however remain indicative and are only followed to varying degrees. Requirements for airborne systems (e.g. Mode S transponders) are presently published on a national basis, with associated overheads and risks of inconsistency ANSPs and aircraft operators should be bound to meeting minimum functional requirements and interoperability standards set at European level by specified dates, according to an agreed Master Plan In a number of cases, new opportunities cannot be exploited due to an absence of corresponding procedures or certification criteria (e.g. ADS-B, SATNAV approaches). Means of compliance and associated operational procedures should be developed and adopted at European level in a timely manner, in keeping with ICAO standards and recommended practices Industry products should be certified for compliance with minimum functional requirements and interoperability standards by authorised bodies These considerations together with point to a changing role for EUROCONTROL in developing systems. 8.9 ECAC institutional strategy The ECAC Institutional Strategy adopted at MATSE 5 in 1997 contains provisions for common systems and projects (appendix 8). There has been little follow-up other than programmes like ARTAS and POEMS. This part of the strategy should be either applied or reviewed in light of recent developments, e.g. joint procurements PRR 5 - Investment

69 8.10 Conclusions Information disclosure requirements should be extended to include data on capital expenditure broken down by major investments (including, e.g., investment in ATM systems, CNS, R&D and infrastructure projects) Ground and airborne ATM systems should meet minimum requirements set at European level for specified areas and periods. ANSPs and aircraft operators should be bound to meeting European minimum functional requirements and interoperability standards by specified dates, according to an agreed Master Plan. This would ensure synchronised implementation and encourage development of compliant products by industry. A central European body should facilitate the adoption of an overall system architecture and Master Plan Means of compliance and associated operational procedures should be developed and adopted at European level in a timely manner, in keeping with ICAO standards and recommended practices. Industry products should be certified for compliance with standards and minimum requirements The new ATM systems should be based on standardised industry products to the maximum extent possible ATM R&D should be organised much more efficiently The provisions of the ECAC Institutional strategy adopted at MATSE 5 in 1997 for common projects and systems (Appendix 8) should be either applied or reviewed. Investment - PRR 5-69

70 Notes on chapter 8 74 The sum of depreciation and interest gives a first approximation of steady state investment levels. 75 Total annual en-route and terminal costs of the European ATM system is approximately 6200 M (see Figure 1). 76 Part of this amount is from public funds (see 8.3.1), which are outside the estimated total 1200 M. 77 The recent Boeing ATM initiative attempts to build on existing results, including from Europe (in particular PHARE), to reach a quantum jump in functionality and performance. 78 The strong safety culture induces a resistance to change in the absence of clear safety cases. 79 But no major cancellations, as opposed to the FAA s Advanced Automation System (AAS). 80 Cost of Swedish S2000 system delay is estimated at 13 M. 81 Source: House of Commons Select Committee on Environment, Transport and Regional Affairs, Third report, 10 January Conversion using exchange rate of 15 April 2002 ( convert.cgi) PRR 5 - Investment

71 9 Fragmentation 9.1 Introduction A wide cost-effectiveness and productivity performance gap between the US and European ATM systems was identified in PRR 4 (ref.10, chapter 8). This fact is now well established. The PRC fully appreciates that there are many different reasons for this difference 82. It is undertaking several studies to identify the dominant explanatory factors and to examine them in more detail. The results of these studies will be available for PRR 6 in Fragmentation has been identified as one of those reasons in PRR 4: the fragmentation of the European ATM system certainly contributes to reducing the overall cost-effectiveness by the multiplication of fixed costs and assets, and higher co-ordination and transaction costs. In addition, the European fragmentation does not allow potential economies of scale to be exploited. Finally, organisational inefficiencies might also result from non-optimal airspace design and civil/military arrangements in Europe Indeed, fragmentation covers a wide range of issues: fragmentation of airspace, ANSPs, ATS units (ACCs, Airport Approach units (APPs), Control tower, civil and military ANS provision, regulation, ATM systems and interfaces. The European ATM Programme (EATMP) and the recent EC Single European Sky initiative [ref.24] both address ways and opportunities to minimise this fragmentation. The level of fragmentation varies between European States The degree of fragmentation has important operational and financial implications for both ANSPs and airspace users. The multiplicity of co-ordination procedures among the various ATS units considerably increases ATCO and pilot workload, contributing negatively to overall ATM safety and productivity. The duplication of ATS units, facilities, and ATC systems increases the fixed costs of ATM, as well as the operating costs comprising overheads, depreciation, maintenance, etc. Because of significant network effects, it is likely that the overall cost of fragmentation at European level is significantly higher than the sum of the costs of fragmentation arising at national level Figure 66 illustrates one aspect of airspace fragmentation: US Air Route Traffic Control Centres (ARTCC) tend to be significantly larger than European ACCs both in terms of traffic and airspace volumes. Because of the presence of fixed costs, larger traffic volumes allow for unit costs to decrease as output (e.g., flight hours controlled or kilometre controlled) increases, until a point is reached where all scale economies are reaped. Whether the US ARTCCs operate at the minimum efficient scale production is still an open question. However, in view of the large dispersion of European ACCs (see Figure 66), it is likely that a significant number of European ACCs do not operate at the minimum efficient scale production. Figure 66: Traffic and size of European and US en-route centres Fragmentation - PRR 5-71

72 9.1.6 The purpose of this chapter is to highlight various initiatives taken by ANSPs and States to reduce the level of fragmentation and to improve operational flexibility and overall efficiency. Some examples illustrate how ANSPs have consolidated their number of ACCs, and/or integrated their ACC and APP(s) into a combined operational unit. Other examples show how civil and military ANSPs have merged to create a civil/military integrated ANSP. Finally, other cases emphasise regional/international initiatives aiming at greater co-operation and more flexible and efficient use of scarce resources. 9.2 National initiatives Various national initiatives have been taken in Europe and overseas to reduce this ATM fragmentation. Such initiatives aim to improve safety, operational flexibility and overall efficiency. A number of recent national initiatives to reduce fragmentation are highlighted below. Germany - DFS: civil/military integration and reduction/consolidation of ATS units In Germany, the integration of military ATS began in February 1994 and was completed by the end of DFS now performs all ATC functions for regional military air traffic in the entire airspace of the Federal Republic of Germany The DFS is now implementing a large scale consolidation project aimed at increasing efficiency. All APP units were moved to the parent ACCs to form fully integrated (and/or co-located) ACC/APP operational units. A reduction in the number of ACCs has been planned. Currently, the DFS operates six centres (Karlsruhe, Berlin, München, Bremen, Düsseldorf, and Langen). Berlin and Düsseldorf are planned to be consolidated in Bremen and Langen, respectively. Switzerland - Skyguide: civil/military integration of ATS units In Switzerland, until 2000, civil and military ANS were provided by two separate ATM organisations (Swisscontrol and the Swiss Air Force), which limited overall flexibility and efficiency of the ANS system. The Swiss authorities decided in 1999 that the civil and military organisations were to be merged and that the necessary institutional ground was to be laid by the end of As of January 2001, Swisscontrol was renamed Skyguide to reflect the new integrated civil/military ATM organisation in Switzerland. On 1 January 2002, Air Force ATCOs were transferred to Skyguide. It is expected that the integration process will be completed by the end of Since it was also decided to merge both ACCs (Geneva and Zurich) into a single UAC, it is expected that the Geneva centre will control all upper airspace under Skyguide responsibility as of 2004, while the lower airspace will be managed by two units in Geneva and Zürich. Romania ROMATSA: Rationalisation of the number of ACCs In line with the ATM Strategy and its new operational concepts, significant decisions for integration and rationalisation have been taken by ROMATSA, actively supported by the State and by the Romanian Civil Aviation Authority (CAA), since it became an autonomous and self-financing company (1991) Until 2001, ROMATSA operated five ACCs (Bucharest, Constanta, Arad, Cluj, and Bacau). The first step in the rationalisation process has been to safely relocate Cluj ACC into Arad ACC and Bacau ACC into Constanta ACC (completed in Nov. 2001). The next step is the operational implementation of the new ATM integrated system allowing for one single ACC (Bucharest) to be operated from three different locations: one main location being the new ACC building in Bucharest (already completed and commissioned) and two secondary, complementary locations in Arad and Constanta, to fully cover the entire airspace while allowing due respect to contingency issues. It is expected that this phase will become operational by the end of 2002, following extensive operational evaluation and approval by the Romanian CAA PRR 5 - Fragmentation

73 9.2.8 Full integration of the ATM system and effective relocation of staff will allow for a further reduction in the secondary ACC locations for the whole Romanian FIR, but such plans will not proceed until the contingency issue is satisfactorily dealt with and until all the operational, technical, social and economical factors have been carefully assessed, including the possibilities of a future wider involvement of ROMATSA at regional level. Australia Airservices Australia: Reduction in the number of ACCs from five to two Airservices Australia (and its predecessor) has managed major changes to the delivery of ATM in Australia. Previously, the Civil Aviation Authority, with more than 7000 staff, managed amongst other things an airspace divided into six Flight Information Regions (FIRs) loosely-based on Australian State boundaries. Air traffic was controlled from five en-route centres, three terminal control units, and 31 towers (see left-hand side of Figure 67) The Australian Advanced Air Traffic System (TAAATS), initiated in 1997 and fully commissioned in 2000, allowed airspace management and ATC operations to be streamlined by consolidating the six FIRs into two, and redesigning sectors to match volumes of air traffic. The Northern FIR (Brisbane) covers the northern half of Australia and oceans to the north and east. The Southern FIR (Melbourne) covers the southern half of the country and large areas of the Indian and Southern oceans (see right-hand Figure 67) Figure 67: Reduction of en-route centres in Australia In parallel to TAAATS, over the period, Airservices implemented a comprehensive Business Transformation project designed to improve the efficiency of its operations to reduce costs, improve staff productivity, reduce charges, and increase value for its shareholder, the Australian Government. Over five years, real cost savings equivalent to 26% of current annual costs have been achieved. ATS operations have been handled safely, staff productivity has significantly increased with a reduction equivalent to 34% in the 1997 work force (see Figure 68). Core staff, i.e. ATCOs, have remained fairly constant throughout the period. Fragmentation - PRR 5-73

74 Figure 68: Airservices staff evolution 9.3 Regional/international initiatives Beyond the creation of the Maastricht UAC in 1972, there are several international initiatives to reduce the fragmentation at a regional level in Europe. These initiatives all originate from the concept of ATS delegation described in Annex 11 of the Chicago Convention [ref.25]. The delegation of the responsibility for providing ATS is a mechanism by which a State entrusts another State with the responsibility of providing ATS over the delegating State s FIR or parts thereof. Such arrangements are encouraged when the management of air traffic independently from national boundaries is deemed to improve the safety, the efficiency and expeditiousness of air traffic for the benefit of both ANSPs and airspace users. EUROCONTROL has developed a Model Agreement on the Delegation of Air Traffic Services [ref.26] to promote a uniform institutional framework covering legal, operational and financial requirements related to ATS delegation in Europe. EU legislation being proposed under the Single European Sky initiative [ref.24] could further reinforce this process. The Franco-Swiss ATS delegation case Among the numerous cases of bilateral ATS delegation in Europe 83, the ATS delegation from France to Switzerland is particularly interesting because of (1) its significant scope, and (2) its location in the core area. Indeed, since 1988, the Geneva ACC has provided ATS in the neighbouring French airspace over an area equivalent to some km 2 (i.e. 40% of the Swiss FIR). In addition, contrary to most bilateral agreements that deal with operational matters only, the Franco-Swiss bilateral agreement includes financial arrangements to allocate costs and revenues among the parties Three recent international initiatives are briefly discussed in the remainder of this chapter: the CEATS programme, the NUAC project, and the Three States Initiative. The CEATS programme The Central European Air Traffic Services (CEATS) programme was created to meet the needs of eight States Austria, Bosnia-Herzegovina, Croatia, the Czech Republic, Hungary, Italy, the Slovak Republic and Slovenia to co-operate in the provision of ATS within their airspace. The programme is being implemented by the EUROCONTROL Agency and CEATS ANSPs. The EUROCONTROL Agency started to provide this support in 1998 by preparing the implementation of the programme. The start of initial operations of CEATS UAC, which will be located in Vienna, is expected in PRR 5 - Fragmentation

75 The NUAC project As of 1 January 2002 a Nordic Upper Area Control Centre (NUAC) project was begun with the objective of establishing an organisation (a limited company or another agreed legal entity) to which ATS provision for the upper airspace above Denmark, Finland, Norway and Sweden can be delegated 84. This will enable the Nordic States to respond to the capacity and regulatory demands arising from the EC Single European Sky initiative, by operating an effective block of airspace and by reducing the number of ACCs in the upper airspace 85. Other explicit long-term objectives of the NUAC Project are: Achieve lower development costs of future ATM systems; Improve interface and inter-operability aspects between ACCs; Enhance flexibility in the use of scarce resources; Introduce standardisation in the Nordic States The NUAC Project has been divided into phases with the first phase, Implementation Planning, envisaged to be finalised in early July The Steering Group for the project comprises the directors of the four ANSPs. To ensure the interests of airspace users and employees, two Reference Groups have been established. From the users side, this comprises representatives from the four national military authorities, SAS and Finnair. The Three States Initiative Under the Three States Initiative (TSI), Albania, the Former Yugoslav Republic of Macedonia, and Greece have signed a Protocol for co-operation in November The aim is to assist in the implementation of co-ordinated programmes and to enable the integration of their air navigation systems at the technical and operational level. 86 The definitive steps are not yet formalised, however they could cover the creation of a single ATC centre concept along the lines of CEATS with a potential location in Thessaloniki and co-ordination of future ATM planning. A common office has been operating in Thessaloniki since January Conclusions Initiatives aimed at reducing civil-military, organisational and airspace fragmentation have demonstrated that improvements can be made to operational and economic efficiency while maintaining safety. Existing and new initiatives should be encouraged. Corresponding provisions already exist under ICAO policy (ATS delegation) and have been expanded by EUROCONTROL for application in Europe, and could be further reinforced by EU Single European Sky legislation when adopted. Fragmentation - PRR 5-75

76 Notes on chapter 9 82 See PRR 4, Other examples include, inter alia, Austria and Slovenia, France and UK, Switzerland and Italy, Switzerland and Germany, Denmark and Iceland. 84 It is interesting to note that a proposal for delegation of ATS regulation to Sweden is being examined as part of the NUAC Project. 85 The working assumption is that the new operating facility is to be based on the new Malmö ACC. 86 Source: Report of the ICAO meeting on normalisation of air navigation situation in the Balkan area, Paris, 21-23/9/ PRR 5 - Fragmentation

77 10 Civil/military issues 10.1 Introduction Following a decision of the Provisional Council in July 2000 based on a PRC recommendation, the Performance Review Unit and EUROCONTROL Agency launched a study to review civil-military co-ordination within Europe. The first phase now being completed, an action plan is being developed by the Agency. This chapter presents the status of ATM civil-military co-ordination in Europe, based mainly on the PRU- EUROCONTROL Agency final report [ref.20] When a congested block of airspace is of simultaneous interest to civil and military users, a compromise must be found. This compromise should bring the best value to society The European Union Single Sky initiative is considered a major step towards better performance in ATM. Well designed and efficient ATM civil-military co-ordination will ensure the continuity of airspace usage across State borders. This is a pre-requisite for the success of the Single European Sky Military activity and military airspace requirements The military activities include: Military traffic flying under GAT (e.g. military air transport); Operational Air Traffic (OAT) within controlled airspace (i.e. military traffic, which does not comply with ICAO rules); and, Military activity in temporary segregated airspace for hazardous day-to-day activities and occasional military exercises Military activity in North Atlantic treaty Organisation (NATO) States [ref.27] decreased significantly in the last 10 years and GAT distance flown increased significantly (see Figure 69). Military activity in non-nato States reduced even more after the disbanding of the Warsaw Pact in Figure 69: Variation of military and GAT traffic Figure 69 shows that the combination of a reduction in military demand and a growth in civil traffic volume, has produced a global increase in the demand to access European airspace. Civil/Military issues - PRR 5-77

78 European military demand is mainly concentrated in a few areas, in particular in a radius of some 300 miles around Belgium. The NATO Tactical Leadership Programme, based in Florennes, generates a load of some 5000 military flying hours [ref.28], concentrated in a period of 15 weeks a year. This area happens to correspond to the highest civil traffic concentration in Europe Even though military demand has decreased, the nature of the military activity has changed: (1) an increased number of unplanned military deployments, (2) an increased need for combined multinational exercises and (3) more requests for larger training areas but generally for shorter periods Military requirements of the European ATM system The European ATM system should be flexible enough to accommodate military requirements. During crisis situations, the system must have: The ability to redesign safely the airspace at short notice The ability to expand ATC capacity at short notice, and to manage GAT traffic flow fluctuations during crisis periods Airspace - a finite resource The location of civil airports and military bases drives airspace requirements. A controlled terminal airspace is necessary around main airports. A military area is necessary close to airbases or where navy exercises usually take place The scarce resource to be shared is space-time. An objective measure is space-time occupancy 87. The average space-time occupancy for military purposes in core Europe during weekdays is normally around 5%. The rest of space-time (95%) is available for GAT operations. However, GAT airspace requirements may conflict with the need for airspace segregation in some parts of the Core Area, during certain periods of the day. Data source: CFMU &AIP Figure 70: Military areas and major GAT traffic flows in the core area Figure 70 shows the interaction between the major GAT flows and the temporary segregated areas (TSAs) in core Europe. Civil requirements for access to airspace is lower at some times of the day. Military activity has minimal impact during these periods. Where airspace is congested, it is necessary to ensure that: 78 - PRR 5 - Civil/Military issues