NATS Fuel Efficiency Metric OA 1161 Version January 2012

Transcription

1 NATS Fuel Efficiency Metric OA 1161 Version January 2012 Prepared by: Chris Nutt

2 This page is intentionally blank

3 Executive Summary During summer 2009 NATS engaged in a series of customer consultation meetings to review the NATS business plan and identify the customer requirements for Control Period 3 (CP3). CP3 represents the third regulatory phase since NATS became a Public- Private Partnership in 2001 and runs from 2011 until Through this consultation process the airline customers voiced a desire to include an additional financial incentive within the regulatory framework for a fuel-performance service quality metric to sit alongside the existing delay measure. A number of highlevel requirements were agreed through customer consultation to test the suitability of candidate metrics estimating fuel performance. These were to: drive behaviours within NATS that lead to a reduction in fuel use for operators accurately reflect the operators fuel performance change from NATS initiatives not provide incentive for actions that may increase fuel use not be unduly affected by factors beyond NATS influence be fair and equitable across NATS customer base be transparent and auditable. As none of the existing metrics were able to meet all of the criteria listed, NATS developed a new metric called the 3-Dimensional Inefficiency Score (abbreviated to 3Di Score) and put this forward as an option for NATS to be incentivised for enabling fuel efficiency. By the start of CP3 in 2011, the 3Di Score was not deemed sufficiently mature to be applied for financial regulation by either NATS or the CAA. This was largely due to the lack of historical monitoring and evidence, using the measure, so a roadmap was agreed to re-assess the situation in August This paper describes the 3Di metric, the key findings of analysis undertaken to fulfil the requirements of the roadmap as well as the NATS proposal put forward as a performance regime to implement financial regulation using the 3Di Score for the remainder of the regulatory period, CP3, as well as the subsequent consultation discussion and changes to the NATS proposal. The key findings and outcomes from the roadmap work is shown below: A metric has been developed called the 3Di score and can be measured for the portion of each flight that occurs in UK domestic airspace. It has been derived using the statistical approach of regression analysis and comprises factors relating to track extension and vertical inefficiency and the interaction between them to produce a score that predicts fuel efficiency for a flight. The scores can range from zero (minimal fuel used) up to around 130. Daily 3Di Scores have been calculated for each day in 2010 and typically range between 15 and 35 with a high degree of variability. Monthly mean 3Di Scores show very low variability which implies a consistent underlying level of structural inefficiency within the UK network. The sampling methodology used to estimate performance between 2006 and 2009 has been shown to be suitable for reviewing past performance. Annual average 3Di Scores have ranged from 28.2 (2006) to 22.7 (2009). Page 1

4 A flight trial has been undertaken between Heathrow and Edinburgh to demonstrate that the 3Di could operationally be reduced. The 3Di Score for the trial flight was 1.4 units (compared to the average on the same route of 22.1). Corresponding fuel savings were identified in the Flight Data Recorder (FDR) information. Between 2006 and 2010 projects delivered by NATS had a limited effect on network-wide environmental performance due to project targets focussing on safety and capacity improvements. A number of future NATS projects are targeting large fuel/co 2 /3Di reductions but are unlikely to deliver during CP3. NATS Operational staff have been consulted to understand ways that 3Di reductions can be targeted during CP3. External factors including traffic levels, traffic mix, weather, and special/exceptional events have all been shown to impact NATS efforts to affect 3Di Scores. The 3Di Score has been reassessed against the agreed metric criteria and deemed suitable for financial regulation with appropriate risks mitigated through implementation. NATS put forward a core proposal of a performance scheme to its regulator and customers in August 2011, with a targeted average annual 3Di score of 25.5 Units of inefficiency until Through the consultation process the CAA tightened this target to 24 Units for 2012 and 2013 with a further stretching target of 23 Units in OA1161 Version January 2012 Prepared by: Chris Nutt Approved by: Steve Hammond Hellen Foster Sam Prince Page 2

5 Table of contents Executive Summary... 1 Table of contents... 3 Glossary Introduction Di Metric Conformance with Metric Requirements Tracking Candidate Metrics Daily 3Di Score during Sampling Accuracy Assessment Annual Environmental Performance Operational Trials Influencing the NATS Annual 3Di Score Responsiveness to NATS Actions Influence of External Factors Developing a Performance Regime Risks on NATS 3Di Performance during CP NATS Proposal Summary Bibliography Appendix 1 Consultation Outcome Page 3

6 Glossary Acronym Definition 3Di 3D Inefficiency AMAN Arrival Manager ANSP Air Navigation Service Provider ATC Air Traffic Control ATM Air Traffic Management BADA Base of Aircraft Data CAA Civil Aviation Authority CFMU Central Flow Management Unit CP3 Control Period 3 FDR Flight Data Recorder IATA International Air Transport Association RFL Requested Flight Level SESAR Single European Sky ATM Research Term Fuel (In)Efficiency Definition A comparison between the estimated fuel burn of a flight against the estimated fuel burn of the reference trajectory for the flight Page 4

7 1 Introduction During summer 2009 NATS engaged in a series of customer consultation meetings to review the NATS business plan and identify the customer requirements for CP3. CP3 represents the third regulatory phase since NATS became a Public-Private Partnership in 2001 and the control period will run from 2011 until 2015 *. Through this consultation process the airline customers identified the importance of fuelefficiency on their operations and the potential impact that ATM can have on the costefficiency of the routes they operate. As a consequence, customers voiced a desire to include an additional financial incentive within the CP3 framework for a fuel-performance service quality metric to sit alongside the existing delay measure. In this way, NATS can strike a balance between fuel-efficiency and delay performance in its policy and decision making, in a similar way to airlines who consider delay and environmental performance in their daily Cost Indexing decisions. A number of high-level requirements were agreed through customer consultation to test the suitability of candidate metrics estimating fuel performance. These were to: drive behaviours within NATS that lead to a reduction in fuel use for operators accurately reflect the operators fuel performance change from NATS initiatives not provide incentive for actions that may increase fuel use not be unduly affected by factors beyond NATS influence be fair and equitable across NATS customer base be transparent and auditable. Together with the candidate metrics agreed with the airline customers, a new measure developed by NATS called the 3Di Score was also considered for suitability. The 3Di Score was developed to combine measures of track extension with vertical inefficiency to predict fuel inefficiency using regression analysis. This paper describes the development of the 3Di Score, how it conforms with the requirements of a metric, looks back at the performance of the metric since 2006 and the tests that NATS has carried out, and summarises the proposal that NATS has agreed with its regulator, the CAA, to be incentivised against during CP3. 2 3Di Metric This section describes the steps taken to derive the 3Di metric equation. To determine fuel inefficiency, the actual fuel use of a flight must be compared to the fuel that would be used for some optimal reference level of performance. For the 3Di Score metric, this reference level of performance, or reference trajectory, is described as an unimpeded climb to the requested cruise level followed by an unimpeded descent, whilst following a direct horizontal track. The principle of the reference trajectory is based around the SESAR user-based trajectory concept. * 01 January December 2014 for en-route operations and 01 April March 2015 for North-Atlantic operations. Page 5

8 Flight Level NATS Fuel Efficiency Metric A number of factors may affect how this trajectory will look for each individual flight. These factors include the aircraft type, the operator business model (and hence the speeds flown), the aircraft weight (considering load factors and cargo carriage), and enroute wind conditions. Of these important factors, only the aircraft type and the requested cruise altitude are filed and stored within the flight plan, the agreed routing documentation shared between the airline and air traffic control. An assumed optimum trajectory was modelled based on aircraft type, requested cruise altitude and route distance to provide a reference quantity of fuel for a flight. The reference trajectories have been modelled using standard BADA aircraft performance data. The BADA database provides typical speed profiles and rates of climb/decent for the most common aircraft types operating within European airspace. We know that different operators utilise the aircraft in different ways and therefore fly at different speeds, but the typical speed profiles from BADA provide a good representation of all operators for a given aircraft type. The reference trajectories follow the great-circle track and cruise at the requested level. International Standard Atmosphere conditions, BADA nominal take-off weight, and still wind conditions have been assumed throughout. An example of such a reference trajectory is shown in Figure 1. Example Reference Trajectory Time (Minutes) Figure 1: An example of a reference trajectory. The estimated reference fuel burn has then been calculated using our in-house model KERMIT and a comparison of the estimated actual fuel burn and reference fuel burn has been made for a sample of 50,000 flights. Clearly, this would include differences resulting from the deviation in the aircraft performance compared to BADA (resulting from airline business model and take-off weight) and the weather conditions encountered. To mitigate these effects, the actual trajectory has been simulated and hence fuel burn simulated - using the same aircraft performance data and atmospheric assumptions as the reference trajectory. KERMIT is a thrust based fuel burn emissions model underpinned by the BADA data suite. Page 6

9 The estimated fuel inefficiency has been calculated for each flight within the data sample and is used as the dependent variable for regression modelling. The explanatory variables required for the regression analysis are track extension and vertical inefficiency, as these are the forms of fuel inefficiency that may be affected by ATM activities both strategically through airspace design and tactically on the day. Track extension takes the form of the proportional excess distance flown, τ, and may be defined as the difference between the actual distance flown, D ACT, and the minimum distance that could have been flown between the same points of entry and exit from UK airspace, D GCD. D D Figure 2 gives a visual representation of the track extension element of the metric. ACT D GCD GCD Figure 2: Track extension diagram for 3Di Score. Vertical inefficiency that results from ATM activities may be simplified to periods of level flight that occur below the requested cruise level. For a given period of level flight, i, the vertical inefficiency, V i, may be summarised as the product of the proportion of flight time spent in level flight and the proportional deviation from its RFL. The fuel burn impact of time spent above the RFL is not well defined since the level may have been requested to either minimise costs relating to delay or those due to fuel. The RFL may also have been requested at a sub-optimal level to avoid weather or an overloaded sector. Hence, for simplicity level offs above the RFL are considered to have zero inefficiency. This concept is formulated below. Within the UKFIR from the first point until the last point of radar contact. Following a geodesic track (i.e. a segment of a great-circle route). Page 7

10 V i ti L l TL 0 i li L l L i Where t i is the duration of the level flight, T is the total duration of the flight, l is the FL during level flight, and L is the reference FL for the flight. Figure 3 gives a visual representation of the vertical elements of the metric. Figure 3: Vertical elements of the 3Di Score. Whilst an aircraft is climbing it is in its most energy intensive phase of flight. During this time potential energy in the form of fuel is converted into kinetic energy (to keep the aircraft airborne) and gravitational potential (altitude). During descent, less fuel is required to keep the aircraft airborne since the gravitational potential can more easily be converted into kinetic energy. Recognising this, the vertical inefficiency may be considered by phase of flight as is shown, where V CL, V CR and V D represent the vertical inefficiency in climb, cruise, and descent respectively. CL i CLIMB CL i CRUISE CL i DESCENT A regression analysis has been conducted using these variables and the 3Di metric developed. The metric takes the form: A B CL C CR D D E CL F CR G D Page 8

11 Where φ is the 3Di score and A, B, C, D, E, F, and G are real constants. More information can be found on the 3Di metric in Air Traffic Control, Business Regulation and CO 2 Emissions (1). 2.1 Conformance with Metric Requirements Other metrics that estimate fuel efficiency of a flight were considered by NATS as options for incentivisation. These metrics were: En-route track extension (developed by Performance Review Unit at EUROCONTROL) Absolute track extension (including the 30Nm region around the airport, differing from en-route track extension) Proportional track extension Absolute fuel use 4D Inefficiency Score (a ratio of aircraft emissions experienced against the estimated emissions of an optimal trajectory flown) 3Di Score A summary of how each of the candidate measures conforms to the requirements to measure an individual flight's inefficiency is shown in Figure 4. Figure 4: Conformance of candidate measures to metric requirements. Whilst it was determined that the only metric that conformed to all of the metric requirements was 3Di, it was agreed between NATS and the CAA that the metric was not deemed sufficiently mature to be applied for financial regulation from the outset of CP3 because of the lack of a track record. Therefore, a roadmap was agreed with the CAA to re-assess the situation in August The relevant steps of the roadmap are shown in Figure 5. Page 9

12 Figure 5: Roadmap detailing steps towards financial regulation of fuel efficiency performance in CP3. Page 10

13 3 Tracking Candidate Metrics 3.1 Daily 3Di Score during 2010 The average 3Di Score for each day in 2010 has been calculated. This is shown in Figure 6. Figure 6: Mean daily 3Di Score for NATS domestic airspace during The daily performance demonstrated substantial variability throughout the year with scores typically ranging between 15 and 35 units **. However, this variability is largely removed when considering performance at a monthly level. The average monthly 3Di Scores for 2010 are shown in Figure 7. ** A 3Di score of zero in mid-april occurred during the period of volcanic ash when no flights were present in UK airspace. Page 11

14 Figure 7: Monthly mean 3Di Score for the UKFIR in The consistency of the monthly performance provides a good indication that the underlying efficiency of the NATS network has remained largely stable during This suggests that daily performance may be broken into two components: The performance of the underlying concept of operations (the airspace, tools, procedures, etc.) The operational response to varying circumstance. 3.2 Sampling Accuracy Assessment Prior to 2010, 3Di scores were not generated for every flight on every day of the year. Due to the computational intensity, a sampling technique was used to calculate the annual average 3Di Score. As the full 2010 dataset was available, the accuracy of the sampling has been studied using the daily 3Di Score data for 2010 to test the accuracy of the annual 3Di score estimates. The error due to sampling for the 2010 data was found to be normally distributed with a mean of zero and a standard deviation of 0.5 units. This provides strong evidence to support the credibility of the annual mean 3Di Scores estimated for using this sampling technique. 3.3 Annual Environmental Performance The average 3Di Score of all NATS revenue earning flights operated during 2010 was The performance in 2010 is shown in relation to past performance in Figure 8. Past performance is shown with error bars representing a 99% confidence interval inferred from the sampling accuracy assessment described in Section 3.2. The average number of revenue earning flights per day is also depicted for reference. Page 12

15 Figure 8: Estimated annual mean 3Di Scores The causes of change in the NATS 3Di Score is described in more detail in Section 5. 4 Operational Trials In July 2010 NATS collaborated with British Airways and the airport operator BAA to operate a normal revenue flight between Heathrow and Edinburgh to test the perfect flight concept and validate the assumptions behind the 3Di Score. Once airborne, the perfect flight concept involves an uninterrupted climb to cruise altitude, a direct route in cruise, followed by an uninterrupted descent to the arrival runway. The route and vertical profile of the flight are shown in Figure 9 and Figure 10 respectively. Page 13

16 Figure 9: Route of Perfect Flight on 10/07/2010. Figure 10: Vertical profile of Perfect Flight (10/07/2010). Page 14

17 The 3Di Score for the perfect flight was 1.4 units. Examination of Figure 10 shows that the perfect flight experienced no periods of level flight below the cruise level, therefore the vertical inefficiency was zero. The 1.4 units of residual 3Di Score can be directly attributed to the Noise Preferential Routes on departure from Heathrow and manoeuvring on to final approach at Edinburgh. The mean 3Di Score for all Heathrow- Edinburgh flights in 2010 was 22.1 units and therefore the perfect flight demonstrated that action can be taken to substantially affect 3Di Scores, with no other flights interacting upon it (or being impacted by the perfect flight). Estimated fuel use data for all Heathrow-Edinburgh flights in 2010 was compared to the estimated fuel use for the perfect flight. This demonstrated a reduction of approximately 300Kg in fuel use whilst airborne. Analysis of corresponding FDR data suggested the benefits of the flight ranged from Kg of fuel and the reduction in 3Di Score shows the impact NATS can have on fuel burn performance in both ATC operation and strategic airspace/route design. 5 Influencing the NATS Annual 3Di Score The roadmap to financial regulation (ref. Figure 5) required NATS ability to affect change in the metric to be demonstrated. This section summarises the findings of analysis undertaken to identify the likely impact of NATS actions and the effect of changes that are outside NATS scope for influence. 5.1 Responsiveness to NATS Actions NATS Action In Section 3.3 the changes in the annual average 3Di Score for UK airspace between 2006 and 2010 were described. During this time a number of operational changes were implemented. In general, these projects were implemented primarily to improve delay or safety performance. While some marked changes in fuel performance have been delivered locally through the Airspace Efficiency Group work at the operational units, there was limited scope for influencing UK-wide performance. As a result, there is insufficient evidence to demonstrably link changes in the UK annual average 3Di Score with the projects delivered between 2006 and Future Changes Through engagement with NATS operational staff during the development of the 3Di Score metric, a number of actions have been identified that could be taken to improve NATS performance. While these concepts have yet to be translated into formal actions, NATS are building a plan to deploy these suggestions over the duration of CP3. Implementation of the metric has also raised awareness of fuel efficiency with NATS operational staff and brought environmental matters into the service thought process. Continuing to raise awareness and providing advice and support to controllers on how to deliver better profiles is likely to help improve NATS fuel efficiency service to its customers. Workshops have been undertaken with operational staff to identify NATS s scope for future impact on the UK annual average 3Di Score. These workshops have provided a strong indication that operational staff believe that a metric that has direct correlations with the profiles they deliver can drive the right operational and management behaviours. However, it was noted that substantial structural changes in the TMA regions would be required to deliver the biggest improvements in NATS performance. Page 15

18 Such changes are planned through the London Terminal Control Airspace Management Programme and the Northern Terminal Control Area development; however the delivery timescales of these projects are unlikely to enable substantial change during CP Influence of External Factors The following sections summarise some of the key external factors that were identified as affecting NATS 3Di Scores. These factors should be considered as risks that may require mitigation as part of the performance regime used to implement financial regulation using 3Di Score Traffic The number of movements a controller is required to manage at any given time is a key component in determining complexity and hence controller workload. Therefore, intuitively it may be expected that the greater the number of movements that need to be handled, the less opportunity the controller has to deliver fuel-efficient profiles. As a result, in general an increase in traffic demand on NATS may be expected to increase the average 3Di Score. The relationship between the annual mean 3Di Score for UK airspace and the average number of movements handled per day is shown in Figure 11. Figure 11: Relationship of annual mean 3Di Score with traffic ( ). This provides some evidence to support the operational expectation of a link between traffic levels and 3Di Score; however the strength of the correlation (R 2 =0.71) and a base of only 5 data points suggests that this link is not conclusive. As a result, this indicates that changing traffic levels could pose a risk to NATS efforts to affect 3Di Scores and should be mitigated appropriately in implementing financial regulation using the 3Di Score. Page 16

19 5.2.2 Flight Category During 2010 a marked difference in performance was identified between different groups of flights operating in UK airspace. Overflights had a daily average 3Di Score of around 4 while domestic flights had an average score of around 25. Flights departing from UK airports to destinations outside of the UK had an average score of around 12 in UK airspace. Upon return these flights had an average score of approximately 44 whilst in UK airspace. The nature of the typical profiles of these flights makes the differences in their 3Di scores intuitively correct. For example, arrivals into UK airspace tend to have a relatively short period of cruise followed by the possibility of some TMA holding due to capacity constraints, resulting in a higher 3Di score; overflights tend to transit UK airspace at or close to their desired cruise level on broadly point to point routings, hence their lower 3Di score. The difference between the performances of each of the groups is demonstrated in Figure 12. Figure 12: Mean 3Di Score for each day of 2010 by Flight Category. This indicates that changes in the proportion of traffic in each flight category could materially affect 3Di performance without any underlying change in NATS performance. Therefore, this poses a risk to NATS efforts to affect 3Di Scores that should be mitigated appropriately in implementing financial regulation using the 3Di Score Weather Weather is an overarching term that covers many factors affecting ATM performance. Broadly these factors may be split into two groups: Discrete events (such as snow, thunderstorms, low visibility, etc.) Factors that vary continuously and always affect the operation (such as wind direction, wind speed, oceanic track positions, etc.) Weather can not only impact the route that aircraft choose to take, but also the efficiency of the runways in use at the airports. This effect can be particularly marked when landing rates are reduced and additional airborne delay is experienced. Page 17

20 The influence of these factors on ATM performance is demonstrated through the application of CFMU flow regulations relating to weather. In 2010 European flights accrued nearly 600,000 minutes of pre-departure delay as a result of weather regulations applied in the UK. This represents over 60% of all pre-departure delay that resulted from flow regulations applied in the UK. Similarly, it may be expected that weather would have a marked impact on 3Di Scores, but the effect should be mitigated through the performance scheme design, outlined in Section Weather Events Analysis of the response of the daily average 3Di Score to weather events such as snow, fog, thunderstorms and high winds failed to identify a clear quantitative link within the 2010 dataset. However, visual inspection of the data showed that such events tended to cause a marked but inconsistent effect. For example, in January 2010 there were a number of days of heavy snow. On the first day of snow the UK average 3Di Score exceeded 36 units (the second highest day recorded in 2010) while the following day a score of 21 units was recorded (well below the yearly average). This may reflect a difference between the tactical reaction on day 1 (which had to reduce landing rates and hence increase airborne holding) and the pre-tactical reaction on day 2 where the reduction in runway capacity resulting from the presence of snow meant that delay was absorbed on the ground rather than in the air. Such inconsistency in the response to such events reflects the difference in the effect on the NATS operation. In particular, the location and time-of-day when such effects occur has a substantial impact on the volume and type of traffic that is affected. Since these effects cannot be predicted, the impact of weather events poses a risk to NATS efforts to affect 3Di performance and should be mitigated appropriately in implementing financial regulation using the 3Di Score Runway Direction Analysis of the 3Di Scores of arrival and departure movements at Heathrow showed a distinct difference in performance based on the predominant runway in use on each day. This is demonstrated in Figure 13. Figure 13: Relationship between daily 3Di Scores of Arrival and Departure movements at Heathrow. Page 18

21 In general, 3Di Scores were on average 12 units higher in easterly conditions than in westerly operations. Consequently, an increase in the need to operate in easterly operations at Heathrow could increase the UK average 3Di Score. Such an outcome would pose a risk to NATS s efforts to affect 3Di Scores; however these effects largely reflect the current airspace structure around Heathrow (which is predominantly designed for westerly wind conditions). As the design of the airspace is within NATS s sphere of influence and the prevailing wind conditions are not expected to considerably alter over the remaining period of CP3, direct action to mitigate this risk is not deemed necessary Airport Scheduling (Airborne Holding) Airborne holding requires additional track miles to be flown at low levels and represents a substantial aspect of NATS s 3Di performance. Airborne holding occurs where the arrival demand for an airport during a particular time frame exceeds the landing rate that can be delivered. While the weather effects described in Section can influence the need for airborne holding, so too does the way in which the airport is scheduled. In the UK over 80% of movements that experience airborne holding are arrivals into Heathrow. Heathrow is an IATA level 3 co-ordinated airport and hence is subject to a fully managed schedule. This process requires that a schedule is agreed between all stakeholders including the airport, the ATC provider, and the airline customers. Presently the schedule is agreed with an expected and accepted 10 minutes of airborne delay on arrival. Since changes to an airport s schedule are primarily a commercial concern for the airport operators, NATS presently has limited influence on scheduling decisions unless there is a clear safety concern. As a result, an airport s decision to change its schedule could have a marked impact on NATS 3Di performance. Despite this, NATS has identified a number of projects (e.g. AMAN and Time-Based Spacing) that could provide a means to absorb such delay without the need to use airborne holding and hence reduce 3Di Scores. Since some of these projects are expected to begin delivering changes to the NATS operation within CP3, the risk of such scheduling changes is not deemed necessary to mitigate at this stage. This is a good example of how having the right metric in place has the potential to drive NATS behaviours to focus on solutions that deliver fuel burn and emissions benefits for our airline customers Events Events (such as industrial action, volcanic ash, sporting events, etc.) have the opportunity to change the dynamics of NATS airspace. This may affect the number of movements, the way in which traffic is presented from neighbouring ANSPs, aircraft routings, and airport performance. Analysis identified that such events pose a similar risk to those identified with weather effects in Section and hence similar mitigation should be sought in implementing financial regulation based on 3Di Score. 6 Developing a Performance Regime The analysis summarised in previous sections was used to develop a performance regime that was proposed and consulted on with external stakeholders. NATS proposal provided a robust mechanism for implementing financial regulation on NATS fuel efficiency performance for CP3, whilst aiming to mitigate the risk of external factors. Page 19

22 6.1 Risks on NATS 3Di Performance during CP3 The following section outlines the primary risks to NATS ability to influence the annual average 3Di Score during CP Traffic Demand and Unanticipated Events In Section 5.2, evidence for a relationship between the number of movements handled by NATS and the annual 3Di Score was identified. As a result, changes in traffic demand during CP3 poses a risk to NATS efforts to reduce 3Di Scores. In particular, traffic forecasts indicate that traffic is anticipated to reach the pre-recession peak levels by Similarly, events which cannot be anticipated, such as weather effects and airport system changes have been shown to impact NATS performance. These impacts do not consistently affect the 3Di Score and so cannot be predicted. As a result, they too pose a risk to NATS efforts to affect 3Di performance during CP Traffic Type Mix In Section 5.2.2, the 3Di Scores from domestic, international departures, international arrivals, and overflight traffic were shown to be markedly different. As a result, changes in the ratio of flights in each of these groups may cause changes in the NATS annual average 3Di Score that do not relate to NATS actions. Whilst these ratios have remained broadly consistent since 2006, should they change during CP3, NATS influence on the change in performance would be diminished Olympic and Paralympic Games (2012) The UK aviation industry is expecting an exceptional period of demand during Summer 2012 as a result of the Olympic and Paralympic Games. This is expected to generate substantial additional traffic. Owing to the complexity of the temporary airspace change that will be implemented to deal these circumstances, NATS is not able to fully anticipate the effect of this period on 3Di Scores. As a result, this period presents a considerable risk that NATS 3Di performance will be adversely affected in Risk Mitigation NATS proposed the following mitigations for the risks outlined in Section : Traffic volume and unanticipated events = Deadband: Given the level of uncertainty seen in the historical data caused by traffic volume and unanticipated events, NATS proposed a deadband around the par value; performance within which would not result in a bonus or a penalty. Traffic mix = Par value modulation: Given potential trends in traffic mix, NATS proposed an automatic function that would change the par value depending on the ratio of key traffic categories. This reflected the expectation that NATS should not benefit purely from a greater proportion of more favourable traffic, and vice versa. Olympics and Paralympics = Exclusion period: Given the uncertain effect of the games on the NATS network, an exclusion period was proposed, during which the metric would not apply. Page 20

23 6.2 NATS Proposal The NATS proposal (ratified by the CAA) followed the precedents set by CAA in determining appropriate performance regimes for the T1, T2, and T3 delay terms already in place for CP3. The generic form of the performance regime is shown graphically in Figure 14. Figure 14: Generic form of the performance regime for fuel-efficiency. NATS put forward the following proposal for incentivisation during CP3, alongside the mitigations outlined in Section 6.1.4: - Par Value: 25.5 Units - Deadband: ± 4.5 Units (range from Units) - Collar: 9 Units - Cap: 54 Units - Financial Gradients: 0.2m per 3Di Unit These numbers were based around the findings from each stage of the roadmap from Section 2.1. This proposal was put to consultation between NATS, CAA and airline customers and a performance regime agreed upon between all stakeholders. 7 Summary This document has summarised the 3Di score metric and shown the analysis that has been undertaken to assess the scope for NATS impact on 3Di scores. In particular, the relationship between daily and annual 3Di performance with external influences and NATS actions have been considered. Such analysis of daily performance has been limited to 2010 data and analysis of annual performance has been limited to Page 21

24 NATS has developed a metric that it believes provides a good measure of the ATM influence on fuel efficiency. The metric and the underlying statistics have been verified by external bodies and the metric has been put forward as a method of incentivising NATS on its environmental performance. The work presented in this document was put forward by NATS as the core proposal to the CAA and customer consultation for the performance regime of the environmental metric from The response to that consultation and subsequent CAA decision is shown in Appendix 1. Bibliography 1. Hammond, Steve, Civil, Jacky and Ross, Matt. Air Traffic Control, Business Regulation and CO 2 Emissions. NATS : Operational Analysis, Awaiting ISBN. Page 22

25 Appendix 1 Consultation Outcome Stakeholder Response to Consultation The core proposal, outlined in Section 6.2, was put forward for stakeholder consultation. The CAA received 10 responses from stakeholders to the core NATS proposal including airlines, IATA and EUROCONTROL. The CAA highlighted that there was strong support for the 3Di metric and desire to introduce the metric as soon as possible from the stakeholders. The stakeholder responses were taken into account by the CAA, alongside the NATS core proposal and their own in-house views before determining the performance regime for the environmental metric during the remainder of CP3. Performance Regime The performance regime put forward for incentivisation for the remainder of CP3 is shown schematically below in Figure 15. Figure 15: Actual performance regime determined by CAA. Other features of the performance regime are: The 2.5 month Olympic and Paralympic period in 2012 is excluded (1 st July 15 th September 2012 inclusive) In 2014, the par value reduces to 23 (and all other components shift down by 1 unit also to keep bonus/penalty gradients the same). NATS can now formulate a plan to improve the 3Di score throughout the remainder of CP3 and aim to reduce the score to such an extent that bonus payments are achieved for environmental performance. Page 23