EUROPEAN ORGANISATION FOR THE SAFETY OF AIR NAVIGATION EUROCONTROL EUROCONTROL EXPERIMENTAL CENTRE CDG REAL-TIME SIMULATION RESULTS EEC Note No. 17/06 Project: Time Based Separation Issued: November 2006 The information contained in this document is the property of the EUROCONTROL Agency and no part should be reproduced in any form without the Agency s permission. The views expressed herein do not necessarily reflect the official views or policy of the Agency.
REPORT DOCUMENTATION PAGE Reference: EEC Note No. 17/06 Originator: EEC APT (Airport Research Area) Sponsor: EATM DAP / APN - European Air Traffic Management -- Directorate ATM Programmes --- Network Organisation and Planning Security Classification: Unclassified Originator (Corporate Author) Name/Location: EUROCONTROL Experimental Centre Centre de Bois des Bordes B.P.15 F 91222 Brétigny-sur-Orge CEDEX FRANCE Telephone: +33 (0)1 69 88 75 00 WEB Site: www.eurocontrol.int Sponsor (Contract Authority) Name/Location: EUROCONTROL Agency 96 rue de la Fusée B-1130 BRUXELLES Telephone: +32 2 729 9011 WEB Site: www.eurocontrol.int TITLE: CDG REAL-TIME SIMULATION RESULTS Authors Anna Wennerberg Marco Gibellini Distribution Statement: Date 11/2006 Project TBS Pages vi + 10 Figures 2 Tables 2 Task No. Sponsor (a) Controlled by: Peter Eriksen (Head of Airport Reseach Area) (b) Special Limitations: None Annexes - Period 2006 References - Descriptors (keywords): Time Based Separation, Wake Vortex, Real-time simulation, Final approach. Abstract: This document resumes the results of the air traffic control Time Based Separation (TBS) real time simulation organised in February and March 2006 by EUROCONTROL. The simulation took place in DSNA/DTI/SDER (Direction des Services de la Navigation Aérienne/ Direction Technique et de l Innovation/ sous Direction Etudes et Recherches appliquées) in Athis-Mons, France. The main conclusions are presented in chapter 7 and the reccomendations in chapter 8.
CDG Real-Time Simulation Results EUROCONTROL TABLE OF CONTENTS LIST OF FIGURES... V LIST OF TABLES... V ACRONYMS AND DEFINITIONS... VI 1. INTRODUCTION...1 2. BACKGROUND...1 3. THE SIMULATION OBJECTIVES...1 4. SIMULATION SETUP...2 4.1. SIMULATION WORKING POSITIONS... 2 4.2. THE EXERCISES... 2 4.3. WIND SCENARIOS... 2 4.4. TRAFFIC SCENARIOS... 3 4.5. PROCEDURES... 3 4.6. TOOLS... 4 4.6.1. Trailing Target Position vector (TTP)...4 4.6.2. Intelligent Time Vector (ITV)...5 4.6.3. Sequencer tool...5 4.6.4. Highlighted labels...5 5. DATA COLLECTION AND INTERPRETATION PRINCIPLES...6 6. MAIN FINDINGS...6 6.1. O1 - SUSTAINING OPTIMAL RUNWAY THROUGHPUT... 6 6.2. O2 MAINTAINING OPERATIONAL SAFETY WHILE USING THE TIME-BASED SEPARATIONS CONCEPT AT ITS NOMINAL/OPTIMAL LEVEL... 7 6.3. O3 OFFERING ACCEPTABLE TOOLS TO THE OPERATORS... 8 6.4. O4 PROPOSING AN OVERALL ENVIRONMENT ALLOWING CONTROLLERS TO OPERATE SMOOTHLY... 8 6.5. O5 - ASSESS/VERIFY THAT EXPERIMENTAL CONDITIONS ARE GOOD ENOUGH (TECHNICAL AND ENVIRONMENTAL MATURITY) TO PRODUCE SIGNIFICANT OPERATIONAL RESULTS... 9 7. CONCLUSIONS...10 8. RECOMMENDATIONS...10 LIST OF FIGURES Figure 1: Example of TTP (Trailing Target Position)... 4 Figure 2: Example of ITV (Intelligent Time Vector)... 5 LIST OF TABLES Table 1: TBS values used during the simulation... 3 Table 2: Non TBS (i.e. ICAO radar separations in NM)... 3 Project TBS EEC Note No. 17/06 v
EUROCONTROL CDG Real-Time Simulation Results ACRONYMS AND DEFINITIONS Abbreviation AMAN APN CDG COOR-INI CWP DAP DSNA DTI EATM EEC E-OCVM FTBS HEAVY HMI IAF ICAO ILS INI ISA ITM ITV LIGHT LOC MEDIUM NM NTBS R/T RTS SDER SEQ STCA TBS TMA TTP TWR VTBS Definition Arrival Manager An advanced Functionality Tool with the purpose of: a) calculate optimum aircraft arrival sequences and times for flights considering preferred flight profiles, aircraft types, performance, weight categories, airspace and runway conditions, inbound flow rate, as well as meteorological (MET) conditions and applying suitable optimisation criteria ; b) Present the planned inbound traffic flow at the controller working positions together with suitably generated advisories for the controller in order to meet the planned arrival sequence and times. Network Organisation and Planning Roissy Charles-de-Gaulle Airport Local CDG acronym for Coordinator Controller Controller Working Position Directorate ATM Programmes Direction des Services de la Navigation Aérienne France Direction Technique et de l Innovation France European Air Traffic Management EUROCONTROL Experimental Centre in Brétigny, France European Operational Concept Validation Methodology Fixed Time Based Separation ICAO wake vortex category for aircraft: weight 136 000 kg or above Human Machine Interface Initial Approach Fix International Civil Aviation Organisation Instrument Landing System Local CDG acronym for Initial Approach Controller Instantaneous Self-Assessment Local CDG acronym for Intermediate Approach Controller Intelligent Time Vector ICAO wake vortex category for aircraft: weight 7000 kg or below Local CDG acronym for Runway Controller ICAO wake vortex category for aircraft: weight between 7000 kg and 136 000 kg Nautical Mile No TBS Radio Transmission Real Time Simulation Sous Direction Etudes et Recherches appliquées France Local CDG acronym for Sequencer Controller Short-Term Conflict Alert An automated system which alerts controllers to potential conflicts between aircraft via the air traffic situation display. Time Based Separation Terminal Control Area A control area normally established at the confluence of ATS routes in the vicinity of one or more major aerodromes. Trailing Target Position vector Tower Control Unit Variable Time Based Separation vi Project TBS - EEC Note No. 17/06
CDG Real-Time Simulation Results EUROCONTROL 1. INTRODUCTION This document resumes the results of the air traffic control Time Based Separation (TBS) real-time simulation organised in February and March 2006 by EUROCONTROL. The simulation took place in DSNA/DTI/SDER (Direction des Services de la Navigation Aérienne/ Direction Technique et de l Innovation/ sous Direction Etudes et Recherches appliquées) in Athis-Mons, France. 2. BACKGROUND The Time Based Separation concept intends to replace distance based with time based separations on final approach. The expected benefit is to regain arrival througput that is lost in strong headwinds because of the lower ground speeds of aircraft. EUROCONTROL Experimental Centre in Brétigny has for some time examined TBS and several studies have led to the conclusion that it is possible to realize such a concept and that some of the major European airports could gain significantly from TBS in terms of punctuality and reduced delays. EUROCONTROL is now conducting validation of the concept following the European Operational Concept Validation Methodology (E-OCVM). This includes a full safety case. So far, most of the validation has been focused on tools and methodology for Air Traffic Controllers and the real time simulation in Athis Mons makes a major contribution to this part of the validation activities. Three weeks of simulations were conducted in an environment and with an HMI similar to CDG terminal area and tower and eleven active CDG air traffic controllers worked at the key positions at final approach and in the tower. The simulation took place during three five-day sessions starting the 8 th February 2006 and ending 10 th March 2006. 3. THE SIMULATION OBJECTIVES The main objectives of the real time simulations were: O1 O2 O3 O4 O5 Sustain optimal runway throughput Maintaining operational safety while using the time-based separations concept at its nominal/optimal level Offering acceptable tools to the controllers Proposing an overall environment allowing controllers to operate smoothly. Assess/verify that experimental conditions are good enough (technical and environmental maturity) to produce useable operational results These objectives were broken down to measurable sub-objectives. These are described in chapter 6. Project TBS - EEC Note No. 17/06 1
EUROCONTROL CDG Real-Time Simulation Results 4. SIMULATION SETUP During the RT simulations active controllers working in the TMA and TWR of the Paris-Charles de Gaulle airport managed different arrival traffic samples under various wind conditions in a familiar working environment. The simulation was done by coupling approach and tower simulators and run joint exercises using the same traffic sample for all of the controller positions. This was the first time for SDER to perform a combined simulation using both tower and approach technical platforms. 4.1. SIMULATION WORKING POSITIONS In order to control all flights from the approach sector until the exit of the runway, five controller positions are needed; these positions are also the evaluated positions: Initial Approach Controller, INI. Intermediate Approach Controller, ITM. Runway Controller, LOC, who is responsible for landing clearances. LOC is also responsible for departing traffic on the parallel inner runway. Sequencer Controller, SEQ, who organises/adapts the sequence proposed by the AMAN. Coordinator Controller, COOR-INI, who coordinates the traffic between the en-route sector and approach sectors. Note that in the simulations SEQ has to continuously update the AMAN sequence according to the real sequence of arriving traffic by coordination with the ITM and INI controllers. Note also that TBS tools are only available at INI, ITM and LOC, whom all are responsible for the spacing of landing aircraft. 4.2. THE EXERCISES An exercise is a combination of wind scenario, traffic scenario, procedures and tools. All aircraft categories, HEAVY, MEDIUM, and LIGHT, were used in the exercises. LIGHT aircraft appears sparsely and only in a few exercises. The duration of each measured exercise is approximately 1 hour. The measurements were made for each arriving flight between the IAF points until the exit of the landing runway. The set of exercises was produced per session week and then repeated each week. The team of controllers also changed every week. Different exercises that used the same wind and traffic sample but once with and once without the TBS vector tool were repeated with the controllers working in the same controller positions both times. 4.3. WIND SCENARIOS It is important to test the TBS concept under several different wind conditions to highlight the potential benefit of TBS capacity recovery whatever the wind conditions are. For simulations, we consider the three following wind scenarios (wind at runway threshold): V1, calm wind condition: wind<10kt. V2, medium headwind condition: 10kt<=wind<=16kt. V3, strong headwind condition: wind>17kt. The simulator for the tower positions did not apply the different wind scenarios. This affects the realism in runway occupancy times and has an effect on numbers of go-arounds. 2 Project TBS - EEC Note No. 17/06
CDG Real-Time Simulation Results EUROCONTROL 4.4. TRAFFIC SCENARIOS Traffic scenarios have been built in order to get a significant load of traffic that can be compared to real traffic at peak hours. The traffic scenarios are based on recorded traffic at Paris Charles-de- Gaulle which has been adapted according to wind scenarios so that the sequence over the IAF points is the same whatever the wind conditions are. This allows assessment of the wind effect or TBS effect on the landing sequence between IAF points until the exit of the runway (measured area). Three traffic scenarios were used: Light traffic: Used for training (-15% of traffic compared to Heavy traffic scenario). Morning peak: This exercise contains 38 % of HEAVY category aircraft. 36-40 acft/hour. Afternoon peak: This exercise contains 9 % of HEAVY aircraft and at least one LIGHT aircraft. 36-40 acft/hour The traffic scenarios do not prerequisite the use of holding stacks but during the simulations, if necessary, controllers could decide to open a stack. Note that during the simulations, missed approaches and go-arounds are always possible (for instance, when an aircraft is not established on ILS localizer or due to a runway incursion). The flight is then normally vectored for a new approach and is re-integrated in the sequence. 4.5. PROCEDURES During experimentations three procedures were tested: 1. Fixed minimum TBS (FTBS), minimum time separation at the runway threshold. 2. Varying TBS (VTBS), FTBS + time buffer guaranties that FTBS are respected at the runway threshold taking into account the eventual catch-up during the last 4NM before the runway threshold according to the aircraft types. 3. Non TBS (NTBS), or ICAO standard distance separations. Table 1: TBS values used during the simulation Leader Follower Heavy Medium Light Heavy 90 110 145 Medium 70 70 125 Light 70 70 70 Table 2: Non TBS (i.e. ICAO radar separations in NM) Leader Follower Heavy Medium Light Heavy 4 5 6 Medium 3 3 5 Light 3 3 3 The tables show the ICAO radar separation distances and their proposed equivalents in time based separations as they were used in the experiments. The time based values were defined by transforming distance to time in zero wind conditions for a typical final approach speed. The time separation figures may still need final adjustment. Project TBS - EEC Note No. 17/06 3
EUROCONTROL CDG Real-Time Simulation Results 4.6. TOOLS 4.6.1. Trailing Target Position vector (TTP) Figure 1: Example of TTP (Trailing Target Position) TTP is a vector support tool which displays on the radar screen the minimum separation between a given pair of arriving aircraft (leader and follower) calculated on time. The TTP vector is directed from the leader aircraft radar symbol and heading backwards. It ends with an arc. During final approach when the aircraft reduces its speed, the distance flown is also reduced, resulting in a gradually shorter length for the TTP. After the first session of exercises the length of TTP was limited to display maximum the length of the ICAO distance separations. Before the descent on the Glide Path, TTP otherwise often became longer than the current ICAO separations and therefore controllers would not apply TBS as a separation norm because it was more penalising than the current distance separations and because it was also difficult to optimise the sequence, i.e. resulting in too big gaps between flights that later could not be reduced. For the purpose of this simulation, when the altitude of the follower was below the altitude of the leader, the ICAO distance separation was displayed as the minimum distance, instead of TBS. This was done to maintain the same safety level as today regarding the risk of encountering wake vortex when a following aircraft flies below the altitude of the leader. This limitation was done automatically by the system. 4 Project TBS - EEC Note No. 17/06
CDG Real-Time Simulation Results EUROCONTROL 4.6.2. Intelligent Time Vector (ITV) Figure 2: Example of ITV (Intelligent Time Vector) ITV shall be seen as a tool which displays the future position of the follower aircraft in TBS seconds. ITV is a vector displayed in front of the follower ending with a small arc. 4.6.3. Sequencer Tool A sequencer tool, similar to the present AMAN, was manually and continuously updated according to the real sequence. It was used to support the calculation of the vector between two aircraft following each other in approach. 4.6.4. Highlighted Labels All labels for HEAVY and LIGHT aircraft used a coloured background in the call sign radar label. See Figure 1. Project TBS - EEC Note No. 17/06 5
EUROCONTROL CDG Real-Time Simulation Results 5. DATA COLLECTION AND INTERPRETATION PRINCIPLES When taking into account the limited number of simulations exercises, the high number of variables and the problems encountered during the simulation runs, the collected data cannot be considered as statistically reliable. From the data analysis we can only identify trends and indications, but we cannot make any definite statements about runway throughput and the reasons behind the differences in the results. 6. MAIN FINDINGS Cells in the table below with no background colour describe goals for the simulation that have been verified using objective measurements: mainly data logged by the simulator software. Cells in the table below with a yellow background describe goals that have been verified using subjective measurements: the questionnaires, the subjective controller workload (ISA), and the debriefings. 6.1. O1 - SUSTAINING OPTIMAL RUNWAY THROUGHPUT O1-1 Compare runway throughput using the NTBS or TBS procedures and analyse the variations. Compare the runway throughput under various wind conditions and analyse the variations. O1-2 Compare runway theoretical maximum capacity and measured throughput under TBS and NTBS. O1-3 Analyse the effect on runway throughput of procedures and wind conditions on the sequence of landing aircraft O1-4 Analyse the controllers perception of time compression in the last NM before the runway threshold and the effect of the tools on arrival sequence optimisation Runway throughput Calm wind Strong head wind No TBS 34.1 (36) 32.6 (33) TBS 36.5 (36) 35.2 (36) The values in brackets represent the theoretical capacity. In theory the runway throughput in calm wind conditions should have been the same. We observe instead a difference. The difference can be attributed to the introduction of the TBS spacing tool (TTP), or the fact that the time separation used in the simulation is less than the translation of the distance separation, or a combination of both. In the case of strong head wind conditions, we observe a higher throughput using TBS than without TBS. In view of these considerations, we cannot univocally identify the reasons behind the results. Controllers have the impression that the use of TTP helps with sequence organisation and optimisation. The TTP displays the minimum separation between two aircraft. The TTP helps the controllers to maintain the required separations, in particular when merging traffic from different flows. 6 Project TBS - EEC Note No. 17/06
CDG Real-Time Simulation Results EUROCONTROL 6.2. O2 MAINTAINING OPERATIONAL SAFETY WHILE USING THE TIME-BASED SEPARATIONS CONCEPT AT ITS NOMINAL/OPTIMAL LEVEL O2-1 Check if the minimum runwayoccupancy time is respected (safety indicator) O2-2 Check the distance separations actually applied to flights over predefined points or at regular intervals (distance infringement) O2-3 Check the time separations actually applied to flights over pre-defined points or at regular intervals (time infringement) O2-4 Analyse the number of incidents (distance/time and vertical infringements, duration of infringement, runway incursions) O2-5 Check whether the occurrence of go-arounds when using the TBS concept compared to NTBS remains insignificant and verify that subsequent occurrences are not related to structural aspects of the TBS concept. O2-6 Analyse the perception of safety (traffic awareness, controller traffic load, controller workload) Not usable as the tower platform didn t simulate the different wind scenarios. It appears that the number of separation infringements is less in TBS than in DBS. One reason could be the help provided by the TTP tool. When using the TTP tool, the minimum separation is clearly visible on the radar screen, it is easier to note when the separation between two aircraft is going to be lost. Not usable data, too many different reasons for go-arounds where identified. Controllers have good traffic-situational awareness which seems not to be impacted by TBS use. Controller workload does not seem to be impacted by the use of TBS, even if the controller's first impression is that his/her workload increases due to the modification of their working methods. The perception of safety is increased when using TBS and the TTP tool. The TTP displays the minimum separation; therefore the controllers do not have to mentally calculate it. The manual update of the AMAN sequence by the SEQ controller may have a negative impact on safety. Indeed, if the ITM changes the order of the landing flights at the last moment, the coordination with the SEQ, even done rapidly, is often done after this change, which induces a change in the pair of flights considered for calculation by the TBS tools. Therefore, in this short period of time, the length calculated by TBS tools may be inconsistent with the real sequence of flights. This is especially critical when a heavy aircraft is involved in the change of sequence Project TBS - EEC Note No. 17/06 7
EUROCONTROL CDG Real-Time Simulation Results 6.3. O3 OFFERING ACCEPTABLE TOOLS TO THE OPERATORS O3-0 Analyse how the controllers perceive the TBS concept. O3-1 Analyse how the controllers perceive the TBS tools (userfriendliness, adapted HMI, level of confidence, etc.) O3-2 Decide if a certain TBS tool is preferred in a given wind condition or at a specific controller position and identify in which circumstances the ITV or TTP is preferred The controllers accepted and understood the TBS concept after its first use. Some controllers preferred the VTBS other the FTBS, according to their way of working. The advantages of VTBS do not justify the additional level of complexity in the tool/algorithm and a bigger change in the controller s working methods. The controllers adapted quickly to the TBS tools and they understood how to make the best use of them. The controllers clearly preferred the TTP to the ITV. The TTP can be used as the norm of separation for the LOC and ITM controller s positions. They would have preferred to have the presently used speed vector instead of the ITV 6.4. O4 PROPOSING AN OVERALL ENVIRONMENT ALLOWING CONTROLLERS TO OPERATE SMOOTHLY O4-1 Evaluate controller workload while using TBS vs. NTBS O4-2 Analyse disturbance to controller working methods (sequence update, go-arounds, stacks, hazardous events, etc.) and the impact on controller efficiency O4-3 Analyse variations in the usage of R/T while using the TBS procedure compared NTBS; Verify that under TBS operations, the R/T load remains acceptable O4-4 Analyse the impact on departing flights when using a closely spaced parallel runway configuration O4-5 Analyse the type of pilot instructions under TBS compared to NTBS (more vectoring? More speed reduction, etc.) and analyse whether the current phraseology is sufficient. The controller workload showed very small differences between TBS and NTBS. Results indicate that globally, the disturbance of controller working method is well accepted. This disturbance is mainly linked to the update of the AMAN sequence, application of time separations that are sometimes lower than the ICAO standard and more focus on final approach to optimise the sequence (flights closer to each other on final approach, possibly increasing controller stress levels). TBS in itself only slightly affect their working method. The controllers had the impression that the R/T communication usage was equivalent. Note that the introduction of TBS requires a change of phraseology (replacing distance separations by time separations when giving traffic information). This is a specific issue for CDG. The numbers of instructions transmitted to the pilots is equivalent and the nature of the instruction is similar in all exercises. Phraseology needs to be adapted for the information phrases that mention the separation (e.g. you are 60 seconds behind instead of you are 3 miles behind ). 8 Project TBS - EEC Note No. 17/06
CDG Real-Time Simulation Results EUROCONTROL O4-6 Analyse the global perception of controller efficiency using TSB compared to NTBS. O4-7 Analyse the method used to update the AMAN sequence. O4-8 Compare controller situational awareness under TBS and NTBS. The controllers feel that they are more efficient when using TBS because of an improved optimisation of the landing sequence. At Charles de Gaulle this is probably mainly due to the introduction of a spacing tool (TTP) and not entirely related to time based separations. The sequence was updated manually. This is a concern as the correct sequence is the basis for the separation algorithm. An incorrect sequence could lead to an incorrect display of the vector length. TBS has no impact on situational awareness when compared to NTBS. 6.5. O5 - ASSESS/VERIFY THAT EXPERIMENTAL CONDITIONS ARE GOOD ENOUGH (TECHNICAL AND ENVIRONMENTAL MATURITY) TO PRODUCE SIGNIFICANT OPERATIONAL RESULTS O5-1 O5-1: Analyse possible skewing of the results (external events which could affect the measures without freezing the platform ) O5-2 O5-2: Analyse the robustness of the platform (freezing, failures, etc.) O5-3 O5-3: Check if experimental conditions resemble the real-life situation O5-4 O5-4: Check the understanding of the concept O5-5 O5-5: Check the level of training The integration of the en-route and the ground simulators (it was done for the first time) caused a number of problems; therefore some of the measurements are not usable. The ground simulator did not simulate the wind. During the four weeks of the simulations the platform has been reliable and it was possible to run and record all but one planned exercise. The simulation environment - the traffic samples, the wind scenarios, and CWP provided a realistic representation of the CDG operational room. Thanks to the appropriate training and the usability of tools, controllers have a good understanding of the TBS concept and have integrated TBS tools very quickly in their working methods (within a few exercises). Project TBS - EEC Note No. 17/06 9
EUROCONTROL CDG Real-Time Simulation Results 7. CONCLUSIONS The main conclusions and findings of this simulations are: The Time Based Separation concept was well received by the controllers. The controllers accepted the Trailing Target Position tool (see picture) better than the Intelligent Time Vector (see EEC note). The controllers understood the changes needed in their working methods. The controller workload is unchanged. The voice communication and the number of orders issued by the controllers are very similar to current operations. A slight gain in throughput was measured. Time based separation is an element of this, but other factors, such as the use of a spacing tool, may also have contributed. Improvements in throughput regardless of wind or the application of time based separation can be obtained by the introduction of a spacing tool. This effect probably varies from airport to airport depending on working methods and the amount of traffic. 8. RECOMMENDATIONS The main recommendations resulting from the simulation are: Refine and test the time based separation values for different aircraft categories using aircraft simulators, taking wind into account. Further investigate the transition between distance based separation (en-route and approach sectors) and time based separation (final approach sectors). Validate the Time Based Separation values and algorithms. Define the requirements of the sequencer tool and if possible introduce automatic update of the sequence (according to electronic strips placed on the controller screen, for example). Investigate the need to adapt Short Term Conflict Alert. Investigate the requirements for adaptation of current phraseology when giving traffic information. Further investigate the correlation between the theoretical and real recovery of capacity in different wind scenarios when using time based separation. In future experiments, distinguish between the contributions to throughput of time based separation and the spacing tool. To use fast time and/or simplified (i.e. with fewer variables in each scenario) real time simulations to further validate the concept. 10 Project TBS - EEC Note No. 17/06