AIRPORT CAPACITY IMBALANCE - Eurocontrol · 2020. 12. 11. · The chance for demand/capacity...
Transcript of AIRPORT CAPACITY IMBALANCE - Eurocontrol · 2020. 12. 11. · The chance for demand/capacity...
PERFORMANCE REVIEW COMMISSION EUROCONTROL, Rue de la Fusée 96, 1130 Brussels, Belgium
TECHNICAL NOTE
07-10-2020
AIRPORT CAPACITY IMBALANCE
STUDY
PERFORMANCE REVIEW COMMISSION
1
Edition History
The following table records the complete history of the successive editions of the present document.
Edition No. Edition Validity Date Reason
0.1 30-07-2020 Initial draft published for consultation
0.2 19-08-2020 Correction of time period used for the study. Clarification of methodology followed for identification of the runway configurations and their share.
0.3 29-09-2020 Reviewed text based on consultation feedback.
0.4 06-10-2020 Runway identification labels added to airport layouts.
2
Contents I. List of Figures .................................................................................................................................. 5
II. List of Tables ................................................................................................................................... 5
1. PURPOSE OF THE STUDY ................................................................................................................ 6
2. BACKGROUND ................................................................................................................................ 6
2.1. THE INTERDEPENDENCY TRIANGLE .......................................................................................... 7
2.2. CAPACITY ANALYSIS ................................................................................................................... 7
2.3. AIRPORT OPERATIONAL PERFORMANCE.................................................................................. 9
2.4. RUNWAY SYSTEM CONFIGURATION ......................................................................................... 9
3. APPROACH .................................................................................................................................... 11
3.1. AVAILABLE DATA ...................................................................................................................... 11
3.2. ANALYSIS .................................................................................................................................. 12
4. RESULTS ........................................................................................................................................ 17
5. CONCLUSIONS .............................................................................................................................. 24
6. REFERENCES ................................................................................................................................. 26
7. ANNEX I: RESULTS SUMMARY ............................................................................................................ 27
1. EBBR (BRU) – Brussels ............................................................................................................... 28
2. EBCI (CRL) - Charleroi ................................................................................................................ 29
3. EDDB (SXF) - Berlin/ Schoenefeld ............................................................................................. 30
4. EDDC (DRS) – Dresden .............................................................................................................. 31
5. EDDE (ERF) – Erfurt ................................................................................................................... 32
6. EDDF (FRA) - Frankfurt .............................................................................................................. 33
7. EDDG (FMO) - Muenster-Osnabrueck ....................................................................................... 34
8. EDDH (HAM) - Hamburg ............................................................................................................ 35
9. EDDK (CGN) - Cologne-Bonn ..................................................................................................... 36
10. EDDL (DUS) - Dusseldorf ....................................................................................................... 37
11. EDDM (MUC) - Munich ......................................................................................................... 38
12. EDDN (NUE) - Nuremberg ..................................................................................................... 39
13. EDDP (LEJ) - Leipzig-Halle ...................................................................................................... 40
14. EDDR (SCR) - Saarbruecken ................................................................................................... 41
15. EDDS (STR) - Stuttgart ........................................................................................................... 42
16. EDDT (TXL) - Berlin/ Tegel ..................................................................................................... 43
17. EDDV (HAJ) - Hanover ........................................................................................................... 44
18. EDDW (BRE) - Bremen ........................................................................................................... 45
19. EETN (TLL) - Tallinn ................................................................................................................ 46
20. EFHK (HEL) - Helsinki/ Vantaa ............................................................................................... 47
3
21. EGBB (BHX) - Birmingham ..................................................................................................... 48
22. EGCC (MAN) - Manchester .................................................................................................... 49
23. EGGD (BRS) - Bristol .............................................................................................................. 50
24. EGGW (LTN) - London/ Luton................................................................................................ 51
25. EGKK (LGW) - London/ Gatwick ............................................................................................ 52
26. EGLC (LCY) - London/ City ..................................................................................................... 53
27. EGLL (LHR) - London/ Heathrow ........................................................................................... 54
28. EGNT (NCL) - Newcastle ........................................................................................................ 55
29. EGPD (ABZ) - Aberdeen ......................................................................................................... 56
30. EGPF (GLA) - Glasgow ............................................................................................................ 57
31. EGPH (EDI) - Edinburgh ......................................................................................................... 58
32. EGSS (STN) - London/ Stansted ............................................................................................. 59
33. EHAM (AMS) - Amsterdam/ Schiphol ................................................................................... 60
34. EICK (ORK) - Cork ................................................................................................................... 61
35. EIDW (DUB) - Dublin ............................................................................................................. 62
36. EKCH (CPH) - Copenhagen/ Kastrup ...................................................................................... 63
37. ELLX (LUX) - Luxembourg ...................................................................................................... 64
38. ENGM (OSL) - Oslo/ Gardermoen ......................................................................................... 65
39. EPWA (WAW) - Warszawa/ Chopina .................................................................................... 66
40. ESGG (GOT) - Göteborg ......................................................................................................... 67
41. ESSA (ARN) - Stockholm/ Arlanda ......................................................................................... 68
42. EVRA (RIX) - Riga ................................................................................................................... 69
43. EYVI (VNO) - Vilnius ............................................................................................................... 70
44. GCFV (FUE) - Fuerteventura .................................................................................................. 71
45. GCLP (LPA) - Gran Canaria ..................................................................................................... 72
46. GCRR (ACE) - Lanzarote ......................................................................................................... 73
47. GCTS (TFS) - Tenerife Sur/Reina Sofia ................................................................................... 74
48. GCXO (TFN) - Tenerife North ................................................................................................. 75
49. LBSF (SOF) - Sofia .................................................................................................................. 76
50. LCLK (LCA) - Larnaca .............................................................................................................. 77
51. LDZA (ZAG) - Zagreb .............................................................................................................. 78
52. LEAL (ALC) - Alicante ............................................................................................................. 79
53. LEBB (BIO) - Bilbao ................................................................................................................ 80
54. LEBL (BCN) - Barcelona .......................................................................................................... 81
55. LEIB (IBZ) - Ibiza ..................................................................................................................... 82
56. LEMD (MAD) - Madrid/ Barajas ............................................................................................ 83
4
57. LEMG (AGP) - Málaga ............................................................................................................ 84
58. LEPA (PMI) - Palma de Mallorca ............................................................................................ 85
59. LEVC (VLC) - Valencia ............................................................................................................ 86
60. LEZL (SVQ) - Sevilla ................................................................................................................ 87
61. LFBO (TLS) - Toulouse-Blagnac .............................................................................................. 88
62. LFLL (LYS) - Lyon-Saint-Exupéry ............................................................................................. 89
63. LFML (MRS) - Marseille-Provence ......................................................................................... 90
64. LFMN (NCE) - Nice-Côte d’Azur ............................................................................................. 91
65. LFPG (CDG) - Paris-Charles-de-Gaulle ................................................................................... 92
66. LFPO (ORY) - Paris-Orly ......................................................................................................... 93
67. LFSB (BSL) - Bâle-Mulhouse................................................................................................... 94
68. LGAV (ATH) - Athens ............................................................................................................. 95
69. LHBP (BUD) - Budapest/ Ferihegy ......................................................................................... 96
70. LICC (CTA) - Catania ............................................................................................................... 97
71. LIMC (MXP) - Milan/ Malpensa ............................................................................................. 98
72. LIME (BGY) - Bergamo ........................................................................................................... 99
73. LIMF (TRN) - Torino Caselle ................................................................................................. 100
74. LIML (LIN) - Milan/ Linate.................................................................................................... 101
75. LIPE (BLQ) - Bologna ............................................................................................................ 102
76. LIPZ (VCE) - Venice .............................................................................................................. 103
77. LIRA (CIA) - Rome/Ciampino ............................................................................................... 104
78. LIRF (FCO) - Rome/Fiumicino .............................................................................................. 105
79. LIRN (NAP) - Naples ............................................................................................................. 106
80. LJLJ (LJU) - Ljubljana ............................................................................................................ 107
81. LKPR (PRG) - Prague ............................................................................................................ 108
82. LMML (MLA) - Malta ........................................................................................................... 109
83. LOWW (VIE) - Vienna .......................................................................................................... 110
84. LPFR (FAO) - Faro ................................................................................................................ 111
85. LPPR (OPO) - Porto .............................................................................................................. 112
86. LPPT (LIS) - Lisbon ............................................................................................................... 113
87. LROP (OTP) - Bucharest/ Otopeni ....................................................................................... 114
88. LSGG (GVA) - Geneva .......................................................................................................... 115
89. LSZH (ZRH) - Zürich .............................................................................................................. 116
90. LZIB (BTS) - Bratislava .......................................................................................................... 117
5
I. List of Figures
Figure 1: The interdependency triangle .................................................................................................. 7
Figure 2: Airport capacity planning phases. ............................................................................................ 8
Figure 3: Steps of the analysis. .............................................................................................................. 12
Figure 4: Difference in peak service rate and share for representative runway configurations .......... 18
Figure 5: Airport Capacity Resilience for the 90 airports analysed in this study .................................. 20
Figure 6: Runway configurations’ impact on peak service rate and performance ............................... 22
Figure 7: Runway configurations’ impact on peak service rate and performance ............................... 23
II. List of Tables
Table 1: Identification of runway system configuration for each time interval. .................................. 13
Table 2: Calculation of peak throughput. ............................................................................................. 14
Table 3: Representative runway system configurations and peak service rates. ................................. 14
6
1. PURPOSE OF THE STUDY
Available runways at a given airport, together with their relative disposition, dimensions,
navigational aids, ATC procedures, environmental constraints and local meteorological
conditions (wind rouse, visibility,…) can result in very different operating conditions, with a
direct impact on the capacity of the runway system and the operational performance.
This study intends to evaluate the potential imbalance between these operating conditions
in terms of capacity and performance, through the identification of runway configurations
and the analysis of the difference in capacity and performance depending on these
configurations, associated with the probability of each configuration.
There are therefore 3 main elements to investigate:
- Capacity per runway configuration.
- Operational performance associated to the runway configuration.
- Runway configuration probability.
2. BACKGROUND
The airport capacity declaration and associated scheduling process entails high complexity as
it is influenced by many different factors.
Airport Airside Capacity refers to the ability of the airport runway/taxiway/apron system to
handle a given demand of flights within a specified time period, incurring an acceptable level
of delay (to be determined by the airport stakeholders). It is defined by the International Civil
Aviation Organisation (ICAO) as the ‘number of movements per unit of time that can be
accepted during different meteorological conditions’ [1], whilst Airport Council International
defines it in terms of ‘maximum aircraft movements per hour assuming average delay of no
more than four minutes, or such other number of delay minutes as the airport may set’ [3].
Airport capacity is a combination of the available infrastructures, the existing ATM/ANS
systems and the capabilities of human actors. Investments in the runway system
infrastructure are usually the most expensive, so the capacity at the apron, taxiway system
and terminal should always be adapted to get the most out of the runway system, being the
determining factor for overall capacity. Additionally, new technologies and optimization in
aircraft spacing and sequencing could result in an airport capacity growth when accounting
for the operational side.
Among all factors affecting the runway system capacity, the runway layout and configuration
is considered as the most relevant. While some airport layouts allow for very similar operation
conditions in one runway configuration or another, resulting in equal or similar capacities for
all possible runway configurations, other layouts do not offer that “symmetric capacity” as
the different configurations might have very different operating conditions, dependencies
and limitations.
7
2.1. THE INTERDEPENDENCY TRIANGLE
Airport capacity, demand and delay are three elements that influence each other (see Figure
1). Capacity refers to the theoretical traffic density the airport can absorb while the demand
corresponds to the airline scheduled operations in correspondence to the declared capacity.
Both elements are linked by the delay which results from the disproportion between the
demand and the available capacity. If one of them changes, the other 2 elements might be
affected. The operational problem to be addressed by planners and operational decision
makers is how to strike a balance between these elements, e.g. how much delay is acceptable
to accommodate more demand.
Figure 1: The interdependency triangle
In case of a change in the runway configuration that reduces the capacity, different scenarios
are possible:
- If the change is unforeseen, the demand is unlikely to be adapted fast enough to this
reduction in capacity. In this case demand (temporarily) exceeds the new (reduced)
capacity and there will be an impact on the airport performance and related delays.
- If the change is unforeseen but the demand remains below the new (reduced)
capacity, the airport system could cope without an impact on performance.
- If the change is foreseen (e.g. night runway configurations due to noise abatement),
demand should be adapted to account for the available runway system capacity. In
this scenario demand will be managed to the available capacity (e.g. limited number
of slots). The chance for demand/capacity imbalance is kept to a minimum but still can
exist.
Hence, assuming no change in demand, a change in runway configuration might have an
impact on the airport performance.
2.2. CAPACITY ANALYSIS
One of the main difficulties found while developing the present study deals with the
calculation of the airport capacity, as there is no universal method for this calculation.
Although all methods might take into account similar factors, there are different approaches.
Among the elements used in such studies are: structural layout (runway, taxiway, apron,
gates, terminals, local airspace); ATC procedures, environmental impact and economic
factors.
8
The analysis and methodologies will also depend on the time horizon for which the capacity
calculation is being made, serving different purposes in the airport capacity plans, as shown
in Figure 2
Figure 2: Airport capacity planning phases.
An example of capacity calculation methodology would follow these steps:
1. Historical throughput data (track performance & areas to prioritize)
2. Tables & analytical models (high level starting point)
3. Simulation models (fast time and real time)
4. Pareto frontiers (optimum solution)
Nevertheless, as previously mentioned, not all airports do regular capacity studies or follow
the same steps. At the moment the only common information repository including airport
capacity is Eurocontrol’s Airport Corner1 [4] where airports can report their capacities on a
voluntary basis. Ideally, this information is provided per runway configuration. However,
many airports do not provide any information in this regard. Even when the information is
available, the differences between methodologies and recurrence in the capacity studies for
airports across Europe make very difficult to have a consistent capacities database.
Peak Service Rate
When demand is above the operational capacity, the delivered throughput (that is, the
number of movements in a certain period of time) will be an indication of the capacity itself.
This is the principle behind the peak service rate: defined as the percentile 99 of the
throughput per hour, it is a proxy for the airport operational capacity, that is, the highest
sustainable throughput the airport can achieve under optimum conditions, assuming there is
sufficient demand, like during periods of congestion.
The percentile 99 is based on the cumulative distribution of the movements per hour over a
sample time period.
Note that at airports which are never or rarely congested, the peak service rate might be
significantly lower than the peak airport operational capacity. This proxy will be valid only
when the period considered includes enough hours where the demand was in fact above the
1 Airport Corner is an airport-focused data repository developed by Eurocontrol with operational information provided by the airports (as a result of local coordination between Airport Operator and ATC)
OPERATIONALPLANNEDSTRATEGIC
-X years M-18 to D-8 D-7 to D
MACRO-STRATEGIC• Long term assessment• Infrastructure capability
STRATEGIC• Mid term assessment• Airline schedule coordination
meeting
PRE-TACTICAL• Short term assessment• Detailed capacity plan
9
capacity. If an airport is underutilised, the peak service rate will simply detect the peak
demand, but not the capacity.
At airports that are rarely congested, the observed maximum throughput can provide a better
indication of the operational capacity.
Using the peak service rate as proxy allows for a consistent analysis of operational capacity
across all airports based on historical data.
2.3. AIRPORT OPERATIONAL PERFORMANCE
As result of airport demand/capacity balancing, the traffic demand will be ‘pushed’ through the capacity bottlenecks (constraints) at the various planning stages and flight phases. Depending on the phase at which the capacity constraint is known, the balancing will be done in different ways. The runway configuration defines one of these capacity constraints.
Until the day before operations: Some runway configurations are already known (or preselected for certain times of the day) long enough in advance for the schedule to be adapted to them and therefore it should result in no impact on performance.
On the day of operations: If there is the need to change to a less favourable runway configuration than the one initially planned, the demand on the runways is managed differently for arrivals and departures: ‐ For arrivals: A combination of arrival regulations (leading to arrival ATFM delay)
together with queuing in the Arrival Sequencing and Metering Area (leading to ASMA additional time). As regulations will only hold the flights before their take-off, they are not able to arrange the arrival flow with immediate effect. Therefore when there is an unexpected change of configuration, the adjustment of the arrival flow is done through holdings and vectoring in the approach.
‐ For departures: A combination of pre-departure delays (flights held at the stand/parking position, with or without departure regulations and leading to ATC pre-departure delay), and queuing at the runway (leading to additional taxi-out time). The balance between these two measures will also depend on other factors like the need to free parking stands.
In summary, a change in runway configuration creates (potentially) an airport capacity resilience issue which has an impact on the airport performance. This impact on performance can be studied by observing the following indicators: ‐ Additional taxi-out time ‐ Additional ASMA time ‐ ATC pre-departure delay ‐ Arrival ATFM delay
2.4. RUNWAY SYSTEM CONFIGURATION
The preferable runway direction is related to the wind conditions (direction and speed) but
the decision on runway configuration (OPS) in use also depends highly on other factors:
10
- Operational Safety. Aircraft lift is by design influenced and benefitted from
aircraft departing and landing into the wind.
- Demand. Arrival and departure demand play a key role in configurations selection,
especially in high demand situation when high capacity configurations are preferred
to serve incoming traffic.
- Air Navigation Systems. Existence, performance and operational status of the
navigational aids (both ground-based and space based) and instrument approach
procedures serving each runway, as different levels of accessibility derived from
asymmetries in instrument approach performance may significantly influence an
airport’s runway configuration.
- Meteorology (ceiling & visibility + wind gusts). Besides wind speed and direction,
other meteorological conditions are of great importance for runway system
configuration, such as visibility and cloud ceiling and wind gusts that can cause serious
harm to aircraft on its vicinity.
- Noise abatement procedures. Noise abatement procedures are used at most major
airports in order to reduce noise impact on neighbour communities and are normally
active at night and early morning period.
- Inertia (controllers’ preference). Air traffic controllers tend to prefer certain runway
system configurations or to remain in a same configuration in order to avoid changes,
so it has an important factor on configuration selection. These habits or Subtle
Navigation Factors (SNF) have been identified in SESAR projects using Machine
Learning Techniques.
- Time of the day (curfews). The time of the day influences the staff availability and the
range of possible configurations that can be selected.
- Coordination (TMA/airport). Flows in and out the airport need to be coordinated,
especially in multi airport Terminal Manoeuvring Areas.
- Other factors: Unavailability of runways (works in progress, maintenance, snow
removal…) or other systems (runway lighting system and MET system status).
Regulatory authorities may further restrict the use of the runway system to so-called
preferential runway system (PRS). Typically, PRS refers to a subset of the total set of runway
system configurations defined by specific conditions.
Research studies are mainly focused on configuration selection process prediction. These
investigations are based on two types of models: prescriptive & descriptive:
- Prescriptive models: Look for an optimal solution (accounting different factors)
- Descriptive models: Conduct historical data analysis
Examples on configuration prediction models include a data-driven model using discrete
choice modelling framework which computes configuration prediction in every next 15 min
interval, extended to 3h probabilistic forecast. Case studies performed in LGA and SFO
11
airports in USA reveal an accuracy of ≈80% [5] Another example is a decision-tree based
model to predict airport acceptance rate used as a decision support tool in Ground Delay
Programs (GDPs) [6].
3. APPROACH
Given the lack of consistent capacity and runway configuration information, the analysis will
use a data driven descriptive model that focuses on the available data in the Airport Operator
Data Flow (APDF) managed by the PRU. Furthermore, this data source is currently used for
the computation of the required indicators for the performance monitoring, so using the
same source for all areas of the study ensures the alignment between them.
3.1. AVAILABLE DATA
The Airport Operator Data Flow is established for 90 airports (status as April 2020) and it
includes, amongst other extensive data for every flight, the runway time (that is, take off time
for departures and landing time for arrivals) for every movement, the type of movement
(arrival or departure) and the runway used.
The data is provided monthly by the airport operators and integrated in a common database
after data quality checks.
This data allows for an approach that addresses the key elements of this study:
- Identification of runway configuration: based on the type of movement, time at the
runway and runway used.
- Runway configuration probability: understood as the percentage of time each
configuration is in use at each airport during the period being analysed.
- Capacity per runway configuration: understood as the peak service rate of each
configuration and calculated as the 99th percentile of the throughput in 1 hour
intervals with such runway configuration.
- Performance indicators for each movement and associated with the corresponding
runway configuration. The analysis of the impact on arrival ATFM delay will be dealt
with in the next version of this study. Due to data quality issues in the provision of pre-
departure delay information, the analysis of ATC pre-departure delay across most of
the 90 airports is not possible at the moment. Therefore the performance indicators
analysed per runway configuration will be Additional ASMA time and Additional taxi-
out time.
These 90 airports present a wide variety of typologies of airport layout (Annex I: Results
summary includes the results with layouts of each of the 90 airports analysed for further
information):
1 runway
2 runways intersecting
2 runways not parallel
12
2 runways parallel and closely spaced
2 runways parallel and independent
3 or more runways
The study is conducted for the calendar year 2019, covering the time window between 7h
and 22h local time for each day.
3.2. ANALYSIS
The APDF data allows for a post-ops data driven analysis including the runway use. The
objective is to identify the runway system configurations in use based on the historical data,
taking into account that in theory each airport with N runways has 6N possible configurations
(assuming each runway can be operated for arrivals, departures or both and in either
direction) (for example an airport with 2 landing strips could be operated in 36 different
ways).
In parallel, as a proxy for capacity, the peak service rate (that is, the percentile 99 of the hourly
throughput) will be calculated for each of these runway system configurations.
Figure 3: Steps of the analysis.
For each airport the analysis involves the following steps:
1. Establish 15 min time intervals that will be used for identification of the runway system
configuration (c.f. Table 1)
2. Identification, in each time interval, of active runways and type of movement (ARR/DEP)
at each runway (c.f. Table 1), resulting in a detected runway configuration for that 15 min.
time interval.
When for a time interval only a type of movement (e.g. arrivals) has been observed, that
configuration will be completed with a runway for the other type of movement (departures
in this example). That runway will be assigned by analysing the adjacent intervals, or if
necessary, assuming the runway is used in mixed mode. This is generally observed at small
airports with very low demand.
3. Identification of sustained runway configurations for each 15 min. rolling hour, by
checking, for each 15 min time interval (i), configurations in the following 45 min [(ii), (iii),
1. APDF in 15 min time intervals
3. Runway configuration for
each interval. Probability
2. Count ARR&DEP per
runway
5. Calculate imbalance
and resilience
4. Calculate peak service rate and
performance for each runway configuration
13
(iv)]. If the configuration observed in (i) is also observed in at least 2 of the next 3 intervals
[(ii), (iii), (iv)], the configuration is considered sustained and valid (condition TRUE in Table
2) and associated to the rolling hour starting with time interval (i).
Table 1: Identification of runway system configuration for each time interval.
The reason to allow 1 in 4 time intervals to correspond to another configuration is to be
able to accommodate some unexpected movement that does not exactly fit with the
runway configuration (in the example in Table 2, the use of runway 25R for a departure of
a Heavy aircraft, although that was not the standard runway for departures in that hour).
The calculation of the time share for each configuration is done on the basis of these rolling
hours with condition TRUE. To identify typical configurations, a minimum share of 3% of
the analysed time has been considered (that is, if a configuration is not active more than
3% of the period of the study, it is not considered representative). The time window
considered is 7 to 22h local time to discard night operations where the demand is normally
too low to consider the peak service rate as a proxy for capacity and the configurations
that are mainly related to environmental constrains.
ARR:25R-DEP:25L
ARR:02-DEP:07R
AIRPORT CALLSIGN TIME UTC LOCAL HOUR PERIOD ARR or DEP RUNWAY
LEBL AFL2512 29-06-2018 20:46 22 4 ARR 25R
LEBL VLG20GK 29-06-2018 20:47 22 4 ARR 25R
LEBL VLG91QL 29-06-2018 20:49 22 4 ARR 25R
LEBL VLG18JC 29-06-2018 20:50 22 4 ARR 25R
LEBL VLG66YE 29-06-2018 20:52 22 4 ARR 25R
LEBL ABG8326 29-06-2018 20:53 22 4 DEP 25L
LEBL KLM81J 29-06-2018 20:53 22 4 ARR 25R
LEBL WZZ1075 29-06-2018 20:55 22 4 DEP 25L
LEBL RYR6369 29-06-2018 20:55 22 4 ARR 25R
LEBL RYR15HD 29-06-2018 20:56 22 4 DEP 25L
LEBL RYR74BH 29-06-2018 20:56 22 4 ARR 25R
LEBL VLG18MC 29-06-2018 20:58 22 4 ARR 25R
LEBL ORO901 29-06-2018 21:03 23 1 DEP 07R
LEBL BCS8049 29-06-2018 21:05 23 1 DEP 07R
LEBL IBK59V 29-06-2018 21:07 23 1 ARR 02
LEBL VLG39LM 29-06-2018 21:09 23 1 ARR 02
LEBL QTR142 29-06-2018 21:09 23 1 DEP 07R
LEBL WZZ152 29-06-2018 21:11 23 1 ARR 02
LEBL IBK7VY 29-06-2018 21:12 23 1 ARR 02
LEBL UAE188 29-06-2018 21:13 23 1 DEP 07R
LEBL AFR144X 29-06-2018 21:14 23 1 ARR 02
LEBL IBK3ZR 29-06-2018 21:16 23 1 ARR 02
AIRPORT CALLSIGN TIME UTC LOCAL HOUR PERIOD ARR or DEP RUNWAY
LEBL AFL2512 29-06-2018 20:46 22 4 ARR 25R
LEBL VLG20GK 29-06-2018 20:47 22 4 ARR 25R
LEBL VLG91QL 29-06-2018 20:49 22 4 ARR 25R
LEBL VLG18JC 29-06-2018 20:50 22 4 ARR 25R
LEBL VLG66YE 29-06-2018 20:52 22 4 ARR 25R
LEBL ABG8326 29-06-2018 20:53 22 4 DEP 25L
LEBL KLM81J 29-06-2018 20:53 22 4 ARR 25R
LEBL WZZ1075 29-06-2018 20:55 22 4 DEP 25L
LEBL RYR6369 29-06-2018 20:55 22 4 ARR 25R
LEBL RYR15HD 29-06-2018 20:56 22 4 DEP 25L
LEBL RYR74BH 29-06-2018 20:56 22 4 ARR 25R
LEBL VLG18MC 29-06-2018 20:58 22 4 ARR 25R
LEBL ORO901 29-06-2018 21:03 23 1 DEP 07R
LEBL BCS8049 29-06-2018 21:05 23 1 DEP 07R
LEBL IBK59V 29-06-2018 21:07 23 1 ARR 02
LEBL VLG39LM 29-06-2018 21:09 23 1 ARR 02
LEBL QTR142 29-06-2018 21:09 23 1 DEP 07R
LEBL WZZ152 29-06-2018 21:11 23 1 ARR 02
LEBL IBK7VY 29-06-2018 21:12 23 1 ARR 02
LEBL UAE188 29-06-2018 21:13 23 1 DEP 07R
LEBL AFR144X 29-06-2018 21:14 23 1 ARR 02
LEBL IBK3ZR 29-06-2018 21:16 23 1 ARR 02
TIME INTERVAL n
TIME INTERVAL n+1
14
Table 2: Calculation of peak throughput.
4. The analysis of the peak service rate requires first the calculation of the throughput for all
rolling hours. Taking then only the hours with valid configurations (that is, rolling hours
where a configuration is sustained in 3 of 4 consecutive time intervals), we calculate the
percentile 99 (peak service rate) for each configuration with a representative time share
(>3%) (see 3.2 Step 3). The throughput analysis can be split in arrivals, departures or total
(c.f. Table 3).
Table 3: Representative runway system configurations and peak service rates.
5. Once a runway configuration has been assigned to certain interval, each flight in that
interval can be associated with a configuration. As the performance indicators are
calculated per flight, this association will be used for the analysis of performance per
runway configuration, resulting in different additional ASMA and taxi-out times per runway
configuration.
Now that all the components of the study have been analysed, the following indicators can
be calculated:
3.2.1. Airport Capacity Resilience
AIRPORT TIME_SLICE CONFIGURATION FOUR_SLICE ARR_HOURLY_THROUGHPUTDEP_HOURLY_THROUGHPUTTOTAL_HOURLY_THROUGHPUT
LEBL 29-06-2018 20:30 ARR:25R - DEP:25L TRUE 34 31 65
LEBL 29-06-2018 20:45 ARR:25R - DEP:25L TRUE 36 31 67
LEBL 29-06-2018 21:00 ARR:25R - DEP:25L TRUE 34 29 63
LEBL 29-06-2018 21:15 ARR:25R - DEP:25L - DEP:25R FALSE 28 32 60
LEBL 29-06-2018 21:30 ARR:25R - DEP:25L TRUE 28 29 57
LEBL 29-06-2018 21:45 ARR:25R - DEP:25L TRUE 28 29 57
LEBL 29-06-2018 22:00 ARR:25R - DEP:25L TRUE 30 27 57
LEBL 29-06-2018 22:15 ARR:25R - DEP:25L TRUE 27 22 49
LEBL 29-06-2018 22:30 ARR:25R - DEP:25L FALSE 21 18 39
LEBL 29-06-2018 22:45 ARR:25R - DEP:25L FALSE 13 16 29
LEBL 30-06-2018 06:00 DEP:07R FALSE 8 19 27
LEBL 30-06-2018 06:15 ARR:02 - DEP:07R TRUE 11 26 37
LEBL 30-06-2018 06:30 ARR:02 - DEP:07R FALSE 9 34 43
LEBL 30-06-2018 06:45 ARR:02 - DEP:07R FALSE 13 37 50
LEBL 30-06-2018 07:00 ARR:25R - DEP:25L TRUE 13 38 51
LEBL 30-06-2018 07:15 ARR:25R - DEP:25L TRUE 17 35 52
AIRPORT CONFIGURATION PERC99_ARR_HOURLY_THROUGHPUTPERC99_DEP_HOURLY_THROUGHPUTPERC99_TOTAL_HOURLY_THROUGHPUT
LEBL ARR:25R - DEP:25L 37 39 70
LEBL ARR:07L - DEP:07R 38 38 70
LEBL ARR:02 - DEP:07R 29 31 56
LEBL ARR:25L - DEP:25L 26 28 49
15
When the probability of a certain runway configuration and the corresponding peak service
rate have been established, the capacity resilience for a given configuration, conf i, will be
calculated as:
Configuration Capacity Resilienceconf i (%) = 1 − (Probabilityconf i ∗ Capacity Reductionconf i)
Where:
Probabilityconf i = share of time intervals with runway system configuration Conf i
Capacity Reductionconf i =Reference Capacity − P99 TotalConf i
Reference Capacity
Reference Capacity = maxi
(P99 TotalConf i)
Conf i ∶ all those configurations with a probability > 3%
Finally, the airport capacity resilience:
Airport Capacity Resilience = mini
(Configuration Capacity Resilienceconf i)
Understanding that the airport resilience will be the lowest of the resilience for the different
configurations.
When the Airport Capacity Resilience is 1, it means that there is no imbalance in terms of
capacity that can be delivered due to runway configurations (either because all
configurations offer the same capacity or because the probability of a runway configuration
with lower capacity is too low).
3.2.2. Impact on the peak service rate
Taking the previously calculated difference between the reference capacity and the peak
service rate of conf i: (Reference Capacity − P99 TotalConf i), calculated for each of the n
representative configurations, and the probability of such reduction (given by the probability
of the configuration i), the impact on the peak service rate, expressed in absolute number of
movements, can be calculated as the weighted average:
Impact on Peak Service Rate = ∑(Probabilityconf i ∗ (Reference Capacity − P99 TotalConf i))
𝑛
𝑖=1
This indicator informs in absolute terms of how many movements per hour are lost (assuming
enough demand) when taking all runway configurations into account, with respect to an
hypothetical 100% share of the most favourable runway configuration in terms of peak
service rate.
16
3.2.3. Impact on the performance indicators (Additional ASMA and taxi-out times)
In terms of additional ASMA and taxi-out times (c.f. 3.2 ANALYSIS: 5 abovePoint 5), the best
performance corresponds to the lowest additional times (that is, less queuing). Therefore in
this case the reference performance is the minimum, as follows:
Reference Add. ASMA time = mini
(Add. ASMA timeConf i)
And in a similar way to the impact on the peak service rate, the total impact on the additional
ASMA time will be the weighted average of the impact for each configuration:
Impact on Add. ASMA time
= ∑(Probabilityconf i ∗ (Add. ASMA timeConf i − Reference Add. ASMA time))
𝑛
𝑖=1
The same calculation applies for the additional taxi-out time.
17
4. RESULTS
This section provides an overview of the results after applying the described methodology to
the 90 airports under study Annex I: RESULTS SUMMARY provides details for each individual
airport. After applying the described methodology to the 90 airports under study, there are
several aspects to be evaluated. This section provides an overview of the results, see Annex
I: Results summary for further information on each airport.
4.1. Identification of runway configuration and probability
The approach shows good coverage in terms of identifying a valid runway configuration for
each time interval. The minimum share of 3% of the analysed time proves to be a reasonable
threshold to discard non-representative configurations, covering more than 95% of the
operations in 2019 for 73 out of 90 airports. Only four airports had less than 90 % of the
operation covered by these representative runway configurations: Hannover (EDDV; 89%
coverage), Helsinki (EFHK; 83%), Rome Fiumicino (LIRF; 86%) and Amsterdam (EHAM; 73%).
In cases like Amsterdam (EHAM) where more available runways result in numerous
configuration possibilities, the 15 minutes intervals might be too short to detect all runways
in use, or the 3% threshold too restrictive (as many runway configurations also mean reduced
shares for each one of them).
When comparing the identified runway configurations for each airport with the information
registered in the Airport Corner, the methodology proves to find not only the main runway
configurations, but also other operating modes that are not recognised in the Airport Corner
but are actually in use at these airports.
4.2. Capacity per runway configuration
Annex I: Results summary presents the detailed results for each airport including the
identified runway configurations and corresponding peak service rates for each airport,
together with the calculated resilience and impact. As mentioned in previous sections, the
main limitation for the identification of the capacity for each configuration is the absence of
enough demand, which might be the case for many of the smaller airports and even some of
the bigger ones.
Figure 4 illustrates the results for the top 30 airports in Europe (from which data is available
for 27) in terms of peak service rate and probability for each runway configuration. Each
bubble represents a runway configuration with the corresponding peak service rate (vertical
axis) and share of utilization (size of the bubble).
18
Figure 4: Difference in peak service rate and share for representative runway configurations
Within these 27 busier airports, 15 have no significant deviations in this peak service rate for
the main configurations, that is, the peak throughput does not seem that affected by the
runway configurations (the difference is 2 movements or less). On the other hand, there are
some cases like Oslo (ENGM), Helsinki (EFHK), Stockholm (ESSA) and Amsterdam (EHAM)
where the dispersion is much higher, and cases like Manchester, where not only there is a
significant deviation, but also an even distribution of the share amongst the configurations.
The proposed indicators (resilience and impact on peak service rate) will integrate these
different aspects (deviations and share) to be able to assess the imbalance of the operation
as a whole.
The comparison of the observed peak service rate and the maximum total throughput with
the capacities declared in the Airport Corner, for those airports where information is
available, yields mixed results.
At airports like Heathrow (EGLL), Gatwick (EGKK), Frankfurt (EDDF), Munich (EDDM),
Dusseldorf (EDDL), Dublin (EIDW), Warsaw (EPWA), Lisbon (LPPT), Zurich (LSZH) or Athens
(LGAV), the calculated peak service rate during 2019 is very close to the declared total
capacity for each runway configuration.
On the other hand, at other airports like Stockholm (ESSA), Helsinki (EFHK), Copenhagen
(EKCH) Rome (LIRF), Charles de Gaulle (LFPG) or Orly (LFPO), the peak service rate, or
051015202530
Fran
kfu
rt (
FRA
)
Am
ster
dam
(A
MS)
Par
is (
CD
G)
Lon
do
n (
LHR
)
Mad
rid
(M
AD
)
Mu
nic
h (
MU
C)
Bar
celo
na
(BC
N)
Ro
me
(FC
O)
Lon
do
n (
LGW
)
Vie
nn
a (V
IE)
Zuri
ch (
ZRH
)
Co
pen
hag
en (
CP
H)
Osl
o (
OSL
)
Du
blin
(D
UB
)
Mila
n (
MX
P)
Sto
ckh
olm
(A
RN
)
Bru
ssel
s (B
RU
)
Du
esse
ldo
rf (
DU
S)
Par
is O
rly
(OR
Y)
Lisb
on
(LI
S)
Ath
ens
(ATH
)
Pal
ma
(PM
I)
Man
ches
ter
(MA
N)
Lon
do
n (
STN
)
Hel
sin
ki (
HEL
)
War
saw
(W
AW
)
Ber
lin T
ege
l (TX
L)
0
20
40
60
80
100
120
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Mo
vem
ents
/ho
ur
Peak Service Rate Total movements (arr + dep) at the top 30 airports in 2019per runway configuration
Bubble size: share of time of a given RWY configuration in use
19
percentile 99 is too low compared to the declared capacities. In these cases, the maximum
delivered throughput comes closer and might be a better guess for the capacity.
Nevertheless, as mentioned in the section 2.2 CAPACITY ANALYSIS, differences between
methodologies and recurrence in the airport capacity studies in Europe also make difficult a
consistent comparison.
4.3. Airport capacity resilience
Figure 5 represents the Airport Capacity Resilience results obtained for the analysed airports
following the described methodology. As highlighted in the chart, only 1 airport, Bratislava
(LZIB) has a resilience below 80%, and only other three airports, Gran Canaria (GCLP), Málaga
(LEMG) and Nice (LFMN) show a resilience value below 90%.
Although this study does not allow to decipher if the imbalance in the peak service rate is due
to lack of capacity or lack of demand, especially for the less busy airports like these four, the
analysis at airport level (See Annex I: Results summary for further information) allows to
better understand the imbalance at each particular airport, unveiling for example cases in
which the most common configuration is a single runway use (even when multiple runways
are available) that has lower peak service rate with an associated detrimental impact on
performance.
20
Figure 5: Airport Capacity Resilience for the 90 airports analysed in this study
The airport capacity resilience is a relative indicator, that is, the magnitude of the problem is
measured with respect to the reference capacity at the airport. Therefore, an imbalance of 4
movements at Amsterdam (EHAM; reference capacity=109) will have a lower reduction of
50%
60%
70%
80%
90%
100%EB
BR
(B
RU
) -
Bru
sse
ls
EBC
I (C
RL)
- C
har
lero
i
EDD
B (
SXF)
- B
erlin
Sch
oe
nef
eld
EDD
C (
DR
S) -
Dre
sden
EDD
E (E
RF)
- E
rfu
rt
EDD
F (F
RA
) -
Fran
kfu
rt
EDD
G (
FMO
) -
Mu
en
ster
EDD
H (
HA
M)
- H
amb
urg
EDD
K (
CG
N)
- C
olo
gne-
Bo
nn
EDD
L (D
US)
- D
uss
eld
orf
EDD
M (
MU
C)
- M
un
ich
EDD
N (
NU
E) -
Nu
rem
ber
g
EDD
P (
LEJ)
- L
eip
zig-
Hal
le
EDD
R (
SCR
) -
Saar
bru
ecke
n
EDD
S (S
TR)
- St
utt
gart
EDD
T (T
XL)
- B
erl
in T
egel
EDD
V (
HA
J) -
Ha
no
ver
EDD
W (
BR
E) -
Bre
me
n
EETN
(T
LL)
- Ta
llin
n
EFH
K (
HEL
) -
Hel
sin
ki
EGB
B (
BH
X)
- B
irm
ingh
am
EGC
C (
MA
N)
- M
anch
este
r
EGG
D (
BR
S) -
Bri
sto
l
EGG
W (
LTN
) -
Lon
do
n L
uto
n
EGKK
(LG
W)
- Lo
nd
on
Gat
wic
k
EGLC
(LC
Y) -
Lo
nd
on
Cit
y
EGLL
(LH
R)
- Lo
nd
on
Hea
thro
w
EGN
T (N
CL)
- N
ewca
stle
EGPD
(A
BZ)
- A
be
rde
en
EGPF
(G
LA)
- G
lasg
ow
Airport Capacity Resilience
50%
60%
70%
80%
90%
100%
EGPH
(ED
I) -
Ed
inb
urg
h
EGSS
(ST
N)
- Lo
nd
on
Sta
nst
ed
EHA
M (
AM
S) -
Am
ster
dam
EIC
K (
OR
K)
- C
ork
EID
W (
DU
B)
- D
ub
lin
EKC
H (
CP
H)
- C
op
en
hag
en
ELLX
(LU
X)
- Lu
xem
bo
urg
ENG
M (
OSL
) -
Osl
o
EPW
A (
WA
W)
- W
arsz
awa
ESG
G (
GO
T) -
Gö
teb
org
ESSA
(A
RN
) -
Sto
ckh
olm
EVR
A (
RIX
) -
Rig
a
EYV
I (V
NO
) -
Viln
ius
GC
FV (
FUE
) -
Fue
rte
ven
tura
GC
LP (
LPA
) -
Gra
n C
anar
ia
GC
RR
(A
CE)
- L
anza
rote
GC
TS
(TFS
) -
Ten
erif
e Su
r
GC
XO
(T
FN)
- Te
ner
ife
No
rth
LBSF
(SO
F) -
So
fia
LCLK
(LC
A)
- La
rnac
a
LDZA
(ZA
G)
- Za
gre
b
LEA
L (A
LC)
- A
lican
te
LEB
B (
BIO
) -
Bilb
ao
LEB
L (B
CN
) -
Bar
celo
na
LEIB
(IB
Z) -
Ibiz
a
LEM
D (
MA
D)
- M
adri
d
LEM
G (
AG
P)
- M
álag
a
LEP
A (
PM
I) -
Pa
lma
de
Mal
lorc
a
LEV
C (
VLC
) -
Va
len
cia
LEZL
(SV
Q)
- Se
villa
50%
60%
70%
80%
90%
100%
LFB
O (
TLS)
- T
ou
lou
se
LFLL
(LY
S) -
Lyo
n
LFM
L (M
RS)
- M
arse
ille
LFM
N (
NC
E) -
Nic
e
LFP
G (
CD
G)
‐ C
har
les
de
…
LFP
O (
OR
Y) -
Par
is O
rly
LFSB
(B
SL)
- B
âle
-Mu
lho
use
LGA
V (
AT
H)
- A
the
ns
LHB
P (
BU
D)
- B
ud
apes
t
LIC
C (
CTA
) -
Cat
ania
LIM
C (
MX
P)
‐ M
ilan
/…
LIM
E (B
GY)
- B
erg
amo
LIM
F (T
RN
) -
To
rin
o C
ase
lle
LIM
L (L
IN)
- M
ilan
/ Li
nat
e
LIP
E (B
LQ)
- B
olo
gna
LIP
Z (V
CE)
- V
enic
e
LIR
A (
CIA
) -
Ro
me
Cia
mp
ino
LIR
F (F
CO
) -
Ro
me
Fiu
mic
ino
LIR
N (
NA
P)
- N
aple
s
LJLJ
(LJ
U)
- Lj
ub
ljan
a
LKP
R (
PR
G)
- P
ragu
e
LMM
L (M
LA)
- M
alta
LOW
W (
VIE
) -
Vie
nn
a
LPFR
(FA
O)
- Fa
ro
LPP
R (
OP
O)
- P
ort
o
LPP
T (L
IS)
- Li
sbo
n
LRO
P (
OTP
) -
Bu
char
est
LSG
G (
GV
A)
- G
ene
va
LSZH
(ZR
H)
- Zü
rich
LZIB
(B
TS)
- B
rati
slav
a
21
resilience than the same imbalance at Bratislava (LZIB; reference capacity=18). The following
indicators (impact on peak service rate and performance) provide an indication of the issue
in absolute terms.
4.4. Impact on peak service rate and performance
The results of the calculated impact on both peak service rate and performance indicators derived from the use of different runway configurations are presented in Figure 6 and Figure 7. Regarding the impact on the peak service rate, 29 of the analysed airports show total capacity symmetry (impact on peak service rate is zero); at 36 airports the impact is less than 1 movement per hour, and 10 more airports have an impact between 1 and 2 movements. From the remaining 15 airports, the highest impacts are observed at Málaga (LEMG), followed by Nice (LFMN), Oslo (ENGM), Gran Canaria (GCLP), Helsinki (EFHK), Dusseldorf (EDDL) and Bratislava (LZIB). Most of these airports that show the biggest impact on peak service rate in absolute terms, also had the lowest resilience results. Additionally, when observing the impact on performance for these airports, it is noticeable that three of them, Oslo (ENGM), Málaga (LEMG) and Helsinki (EFHK), also suffer important combined impacts on the performance indicators of above 1 minute per flight. As explained before, an unexpected change of configuration can worsen performance in the
form of higher delays and holdings. In this study two performance indicators related to the
management of the arrival and departure flows are analysed separately for the identified
runway configurations (c.f. 3.2.3). Nevertheless, the runway configuration also has a great
influence in the taxi-out and approach routes, where it might perfectly be that one
configuration is more prone to suffer bottlenecks on the taxiway system or conflicting
approaches. In that case, even with equal or similar peak service rates, the performance of
two runway configurations might be very different. This is the case for airports like
Saarbruecken (EDDR), Porto (LPPR), Budapest (LHBP), Leipzig (EDDP), Vienna (LOWW), Zurich
(LSZH) and even Frankfurt (EDDF), all of them with an impact on delays above 1 minute per
flight (additional ASMA and taxi-out combined).
The detailed analysis airport by airport (see Annex I: Results summary) shows as well the peak service rates for arrival and departures for each runway configuration, together with the maximum hourly throughput observed in 2019. The results for arrivals and departures also evidence that some configurations are preferential for departure peaks and others for arrival peaks (e.g. Amsterdam)
22
Figure 6: Runway configurations’ impact on peak service rate and performance
0 2 4 6 8 10 12
GCLP
GCFV
EYVI
EVRA
ESSA
ESGG
EPWA
ENGM
ELLX
EKCH
EIDW
EICK
EHAM
EGSS
EGPH
EGPF
EGPD
EGNT
EGLL
EGLC
EGKK
EGGW
EGGD
EGCC
EGBB
EFHK
EETN
EDDW
EDDV
EDDT
EDDS
EDDR
EDDP
EDDN
EDDM
EDDL
EDDK
EDDH
EDDG
EDDF
EDDE
EDDC
EDDB
EBCI
EBBR
(movements/hour)
Impact Peak Service Rate
00.511.522.5
GCLP
GCFV
EYVI
EVRA
ESSA
ESGG
EPWA
ENGM
ELLX
EKCH
EIDW
EICK
EHAM
EGSS
EGPH
EGPF
EGPD
EGNT
EGLL
EGLC
EGKK
EGGW
EGGD
EGCC
EGBB
EFHK
EETN
EDDW
EDDV
EDDT
EDDS
EDDR
EDDP
EDDN
EDDM
EDDL
EDDK
EDDH
EDDG
EDDF
EDDE
EDDC
EDDB
EBCI
EBBR
(min/flight)
Impact Add. Taxi-Out Time Impact Add. ASMA Time
Impact of runway configurations usage on operational performance
23
Figure 7: Runway configurations’ impact on peak service rate and performance
0 2 4 6 8 10 12
LZIB
LSZH
LSGG
LROP
LPPT
LPPR
LPFR
LOWW
LMML
LKPR
LJLJ
LIRN
LIRF
LIRA
LIPZ
LIPE
LIML
LIMF
LIME
LIMC
LICC
LHBP
LGAV
LFSB
LFPO
LFPG
LFMN
LFML
LFLL
LFBO
LEZL
LEVC
LEPA
LEMG
LEMD
LEIB
LEBL
LEBB
LEAL
LDZA
LCLK
LBSF
GCXO
GCTS
GCRR
(movements/hour)
Impact Peak Service Rate
00.511.522.5
LZIB
LSZH
LSGG
LROP
LPPT
LPPR
LPFR
LOWW
LMML
LKPR
LJLJ
LIRN
LIRF
LIRA
LIPZ
LIPE
LIML
LIMF
LIME
LIMC
LICC
LHBP
LGAV
LFSB
LFPO
LFPG
LFMN
LFML
LFLL
LFBO
LEZL
LEVC
LEPA
LEMG
LEMD
LEIB
LEBL
LEBB
LEAL
LDZA
LCLK
LBSF
GCXO
GCTS
GCRR
(min/flight)
Impact Add. Taxi-Out Time Impact Add. ASMA Time
Impact of runway configurations usage on operational performance
24
5. CONCLUSIONS
This study presents a data-driven analysis of the airport capacity imbalance from a conceptual
definition to its implementation. It analyses the configurations of 90 European airports, based
on one year of data (2019) covering airport movements between 7h and 22h local time, and
the risk of capacity reduction associated to the change from one configuration to another,
together with the impact on throughput and performance. The main conclusions are
summarised as follows:
The followed data-driven approach provides a common framework for airport
configuration determination which ensures that the real airport operation is taken into
account rather than the declared or planned one. This allows an easy-to-maintain
methodology for the identification of the runway system configuration in use at each
airport at each time.
The runway configurations obtained via this data-driven methodology coincide or are very
similar to those published in the Airport Corner for the majority of the airports. That means
that the data-driven detected configurations with a minimum 3% probability are in the
majority of cases the same as the ones declared by the airports.
Regarding the identification of the capacity for each runway configuration, the peak service
rate is in some cases significantly lower than the declared capacities in the Airport Corner
That may be an indication that these airports still have some buffer for traffic increase,
while the others, where the peak service rate is close to the declared capacity, may be
already operating at their maximum capacity.
Resilience values are calculated based on the reference capacity (maximum peak service
rate of all available configurations at the airport), accounting for real operation aspects of
the year of data analysed (2019). This implies that the resilience obtained corresponds to
the time period analysed; whenever operation in the airport changes, resilience values will
also change to adapt to the new situation. Hence, the airport resilience is a dynamic
indicator and its evolution can be monitored.
The daily time-window covered in the analysed goes between 7h to 22h local time, which
ensures curfews (like for noise abatement procedures) are not taken into account in the
analysis since they are normally active in the in-between period. However, in some cases
specific noise abatement configurations might be activated outside of this time window,
with its consequent impact on the resilience calculation, as night configuration imply less
staff and normally also less demand in the airport.
Resilience results suggest there is no drastic decrease in capacity (in relative terms) related
to a runway configuration change at most airports. Whenever there is a significant
decrease in capacity, it occurs for a configuration with a very low probability, as such
limiting the impact on the airport’s resilience.
Most of the 90 airports analysed show very small changes in capacity for the most common
runway configurations, as derived from the Impact on peak service rate. There are some
of them, though, that show significantly less throughput and higher delays.
25
The impact on the additional taxi-out and ASMA times yields very interesting results and
evidences, at some airports, an important imbalance in terms of performance for different
runway configurations. These results highlight configurations that, beyond a potential
capacity issue, have intrinsic bottlenecks in both approach and taxi-out procedures.
The environmental constrains, the runway closures due to works in progress and many
other factors like implementation of new procedures have a clear impact in the identified
runway configurations and performance. The high level results are to be completed by the
analysis per airport and when important imbalances are detected, further breakdown per
month and/or hours of the day can bring additional insight.
In summary, this data driven analysis allows to study the imbalance of the operation
associated to the runway configurations use and its impact on performance, helping in the
identification of operational constraints at airports. This can significantly contribute to the
fine tuning of the process of assessing capacity limits specially delivered by slot coordinators
at the congested airports.
26
6. REFERENCES
[1] ICAO. Doc 9883 “Manual on Global Performance of the Air Navigation System”, 1st
Edition (2009)
[2] Malawko, Jan: “Airport capacity imbalance metrics”. Presentation to the
Performance Review Commission (2018)
[3] ACI. “Guide to Airport Performance Measures”, (2012).
[4] EUROCONTROL Airport Corner: https://www.eurocontrol.int/tool/airport-corner.
[5] Avery, Jacob, and Hamsa Balakrishnan. "Predicting airport runway configuration: A
discrete-choice modeling approach" (2015).
[6] Jones, James C., et al. "Predicting & quantifying risk in airport capacity profile
selection for air traffic management" 14th USA/Europe Air Traffic Management
Research and Development Seminar (ATM2017), Seattle, USA. (2017).
[7] U.S./Europe comparison of air traffic management-related operational performance
for 2015: https://www.eurocontrol.int/publication/useurope-comparison-air-traffic-
management-related-operational-performance-2015.
27
7. ANNEX I: RESULTS SUMMARY
This annex compiles the results obtained for each of the 90 airports analysed, including the following
indicators:
Annual traffic 2019
Number of configurations
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact PSR2 (mov/hour)
Airport Resilience
A graph displaying these results for each configuration:
Configuration identification with active runways for arrivals (ARR)
and departures (DEP)
Probability
Peak Service Rate (P99) Total movements
Maximum Throughput Total movements
Peak Service Rate (P99) Arrivals
Maximum Throughput Arrivals
Peak Service Rate (P99) Departures
Maximum Throughput Departures
Average Additional ASMA time per arrival
Average Additional Taxi-out time per departure
The airport layouts are also included for a better understanding of the configurations and
identification of potential interdependencies in the use of runways.
This comprehensive overview allows to evaluate the main operating modes and differences in the
resulting performance.
2 PSR: Peak service rate
28
1. EBBR (BRU) – Brussels
The analysis of the operation during 2019 at Brussels shows 5 representative configurations, 4 of them
in segregated mode. The most common configuration does not show the highest peak service rate but
it does show the best arrival rate and low additional times compared to the rest. Configuration
ARR:01-DEP:07R (7%) causes the highest delays in both ASMA and taxi-out, with a lower peak service
rate.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact PSR (mov/h)
Airport Resilience
EBBR Brussels 229281 5 0.37 0.40 2.30 97.12%
58% 21% 7% 6% 3%
31
42
3538 37
41
46
39 40 3942
24
3533
31
50
31
36 3633
5760
55
59
54
65 6562
65
58
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0
10
20
30
40
50
60
70
ARR:25L - ARR:25R -DEP:25R
ARR:25L - DEP:25R ARR:01 - DEP:07R ARR:07L - DEP:07R ARR:19 - DEP:25R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EBBR (BRU) - Brussels
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
29
2. EBCI (CRL) - Charleroi
Charleroi is a single runway airport with only two possible operating modes and a clear runway
direction preference. In general terms the performance in both runway configurations is very similar,
and only additional taxi-out when using RWY06 shows an increase compared to when using the
preferential runway direction (RWY24).
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EBCI Charleroi 54763 2 0.03 0.01 0.87 96.23%
87% 13%
13 14
19
17
13 13
18
2322
23
32
28
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
0
5
10
15
20
25
30
35
ARR:24 - DEP:24 ARR:06 - DEP:06
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EBCI (CRL) - Charleroi
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
30
3. EDDB (SXF) - Berlin/ Schoenefeld
Berlin Schoenefeld operates only one runway (07L/25R) in both directions mixed mode, with almost
identical result in performance.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDB Berlin/ Schoenefeld
90231 2 0.02 0.06 0.00 100.00%
71% 29%
14 14
20
18
14 14
19 19
25 25
31
29
0.0
0.5
1.0
1.5
2.0
2.5
0
5
10
15
20
25
30
35
ARR:25R - DEP:25R ARR:07L - DEP:07L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EDDB (SXF) - Berlin/ Schoenefeld
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
31
4. EDDC (DRS) – Dresden
Dresden is a single runway airport, operated in mixed mode in both directions, being RWY22 the
preferential. Performance in terms of peak service rate is identical, and while RWY04 results in longer
additional ASMA times, RWY 22 shows higher additional taxi-out times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDC Dresden 20524 2 0.06 0.03 0.00 100.00%
78%22%
6 6
7 7
6 6
7
8
9 9
13
10
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0
2
4
6
8
10
12
14
ARR:22 - DEP:22 ARR:04 - DEP:04
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EDDC (DRS) - Dresden
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
32
5. EDDE (ERF) – Erfurt
Erfurt operates one runway in mixed mode in both directions, being RWY28 the preferential. The peak
service rate for RWY28 is 2 movements higher than for RWY10, but the maximum throughput is
actually higher for RWY10, which leads to think the imbalance in peak service rate is driven by demand
and not capacity. In terms of performance, it is interesting to observe how the preferential runway
use results in higher additional times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDE Erfurt 4607 2 0.22 0.25 0.53 97.33%
73% 27%
11
9
1213
10 10
1213
20
18
24
26
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
0
5
10
15
20
25
30
ARR:28 - DEP:28 ARR:10 - DEP:10A
dd
itio
nal
tim
es (m
inu
tes/
fligh
t)
Mo
vem
ents
/ho
ur
EDDE (ERF) - Erfurt
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
33
6. EDDF (FRA) - Frankfurt
Frankfurt, the busiest airport in Europe, operates 4 runways, 3 of them in both directions and 1 (18/36)
only as RWY18. This results in many different configuration possibilities, from which the study
identified only 3 as representative, covering 97% of the operation. Out of these 3 configurations, 2 are
identical to those declared in the Airport Corner in 2019, and one is very similar. Peak service rate for
the three configurations is almost the same, resulting in very high resilience and low impact on the
PSR. Regarding performance, configuration ARR:25L-ARR:25R-DEP:18-DEP:25C, used 22% of the time,
shows the considerably worse additional times, especially in the taxi-out phase, with more than 5
min/dep.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDF Frankfurt 513866 3 0.66 0.33 0.58 99.68%
40% 35% 22%
6258 61
68 65 6765 6561
72 7166
110 109 109
121116 119
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0
20
40
60
80
100
120
140
ARR:07L - ARR:07R - DEP:07C -DEP:18
ARR:25C - ARR:25L - ARR:25R -DEP:18 - DEP:25C
ARR:25L - ARR:25R - DEP:18 -DEP:25C
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EDDF (FRA) - Frankfurt
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
34
7. EDDG (FMO) - Muenster-Osnabrueck
Muenster has one runway (07/25) that can only be operated in mixed mode in both directions. PSR is
identical for both configurations and performance is very similar, with a small impact on the additional
ASMA time when using RWY07 (30% share). The big dispersion between the maximum observed
throughput and the PSR signals the airport might be underutilised and therefore the PSR would not
be a good proxy for capacity.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDG Muenster-Osnabrueck 18604 2 0.01 0.03 0.00 100.00%
70% 30%
7 7
1011
15 15
2322
19 19
28
26
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0
5
10
15
20
25
30
ARR:25 - DEP:25 ARR:07 - DEP:07A
dd
itio
nal
tim
es (m
inu
tes/
fligh
t)
Mo
vem
ents
/ho
ur
EDDG (FMO) - Muenster-Osnabrueck
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
35
8. EDDH (HAM) - Hamburg
Hamburg has 2 crossing runways, resulting in 6 identified runway configurations. The PSR is similar for
all configurations (1 or 2 movements difference, so high resilience results), even when using only one
of the runways in mixed mode (ARR:23-DEP:23 or ARR:33-DEP:33). However, single runway
configurations show a clear impact on performance, with much higher additional times. The most
commonly used configuration offers the best results in performance and throughput.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDH Hamburg 149221 6 0.36 0.41 0.39 99.52%
46% 19% 13% 10% 5% 5%
22 22 22 22 21 21
2826 25 25 25 2423 22 22 22 21 21
28 27 27 27
23 23
40 39 40 4038 38
4846
4345
41 41
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
10
20
30
40
50
60
ARR:23 - DEP:33 ARR:15 - DEP:23 ARR:23 - DEP:23 ARR:05 - DEP:33 ARR:33 - DEP:33 ARR:15 - DEP:05A
dd
itio
nal
tim
es (m
inu
tes/
fligh
t)
Mo
vem
en
ts/h
ou
r
EDDH (HAM) - Hamburg
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
36
9. EDDK (CGN) - Cologne-Bonn
Köln has 3 runways, but for almost 80% of the time, the airport operates only one (14L/32R) in single
runway mixed mode, as the other 2 runways are much shorter and might not offer the same
navigational aids. The maximum throughput and PSR is reached though, when using also RWY 24 for
arrivals in addition to 14L in mixed mode. In terms of performance, there is an imbalance in the
additional times between configurations, and different impact depending on the indicator.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDK Cologne-Bonn 140929 3 0.03 0.19 2.00 95.90%
41% 38% 12%
15 1516
19 19 19
16 1618
20 2021
2728
30
3334
35
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0
5
10
15
20
25
30
35
40
ARR:14L - DEP:14L ARR:32R - DEP:32R ARR:14L - ARR:24 - DEP:14L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EDDK (CGN) - Cologne-Bonn
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
37
10. EDDL (DUS) - Dusseldorf
Dusseldorf has two parallel runways that are sometimes used in segregated mode around 30 % of the
time. However 57% of the operation is handled in a single runway mixed mode, which results in much
lower PSR and maximum throughput, and also less efficient taxi-out times. Operation of both RWY
simultaneously is restricted to 56 hours per week due to court ruling, which results in a resilience that
is consequently lower than at other airports, and also a higher impact on the additional taxi-out times.
It is worth mentioning that works north of runway 05L/23R started in Summer 2019 impacting the
normal operation, as arrivals could not use this runway.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDL Dusseldorf 225541 5 0.41 0.15 6.37 91.30%
48% 25% 9% 5% 5%
28
35
28
3229
3438
31
37
31
25
32
25
31
2629
37
27
3228
45
54
44
55
4751
57
46
59
48
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0
10
20
30
40
50
60
70
ARR:23L - DEP:23L ARR:23R - DEP:23L ARR:05R - DEP:05R ARR:05L - DEP:05R ARR:05R - DEP:05R -DEP:23L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EDDL (DUS) - Dusseldorf
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
38
11. EDDM (MUC) - Munich
Munich offers a perfect symmetrical layout with 2 parallel independent runways, both of them used
in mixed mode, with RWYs 26L/R used 61% of the time. The symmetry also applies to the results in
terms of PSR and maximum throughputs leading to high resilience. The observed performance,
although similar, is slightly better when using RWYs 08L/R
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDM Munich 414222 2 0.23 0.11 0.37 99.59%
61% 37%
58 59
67 66
56 56
64 62
90 89
97 96
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0
20
40
60
80
100
120
ARR:26L - ARR:26R - DEP:26L - DEP:26R ARR:08L - ARR:08R - DEP:08L - DEP:08R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EDDM (MUC) - Munich
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
39
12. EDDN (NUE) - Nuremberg
Nuremberg is a single runway airport, used in mixed mode on both directions and with a big dispersion
between the maximum observed throughput and the PSR, which signals the airport might be
underutilised and therefore the PSR not be a good indication for capacity. In terms of performance,
there are almost no differences between runway configurations.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDN Nuremberg 49207 2 0.02 0.01 1.94 92.80%
65% 35%
13 14
19
32
1214
22 2324
27
40
43
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0
5
10
15
20
25
30
35
40
45
50
ARR:28 - DEP:28 ARR:10 - DEP:10
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EDDN (NUE) - Nuremberg
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
40
13. EDDP (LEJ) - Leipzig-Halle
Leipzig has 2 parallel runways and not that many movements for that infrastructure. The analysis
indicates that for 82% of the time, the airport is operated in a single runway mixed mode, and only
10% in segregated mode. In fact, significant lower throughput is observed for the segregated modes
and also significant worse additional taxi-out times for conf. ARR:08L-DEP:08R.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDP Leipzig-Halle 75413 6 1.02 0.27 1.31 96.68%
32% 26% 21% 6% 4% 3%
9
1110
87
10
16
1213
9
7
11
9
1110
810 10
11
1514
9
12
10
17
19 19
12
15
17
26
22
25
13
15
20
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0
5
10
15
20
25
30
ARR:26R -DEP:26R
ARR:26L -DEP:26L
ARR:08R -DEP:08R
ARR:08L -DEP:08R
ARR:26L -DEP:26R
ARR:08L -DEP:08L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EDDP (LEJ) - Leipzig-Halle
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
41
14. EDDR (SCR) - Saarbruecken
Saarbruecken is a single runway airport, operated in mixed mode in both directions, being RWY27 the
preferential. Performance in terms of peak service rate is identical, but the performance for the
preferential runway in terms of additional times is considerably worse, resulting in a high impact on
both additional taxi-out and ASMA times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDR Saarbruecken 7978 2 0.50 0.59 0.00 100.00%
73%26%
4 4
6
55
4
6
5
7 7
8
7
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0
1
2
3
4
5
6
7
8
9
ARR:27 - DEP:27 ARR:09 - DEP:09
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EDDR (SCR) - Saarbruecken
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
42
15. EDDS (STR) - Stuttgart
Stuttgart operates only one runway (07/25) in both directions mixed mode, with almost identical
result in performance. There is however a significant difference between the PSR and the maximum
throughput, which questions the validity of PSR as indication for capacity
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDS Stuttgart 132609 2 0.01 0.07 0.62 98.40%
62% 37%
23 24
2931
22 23
2830
38 39
4749
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
10
20
30
40
50
60
ARR:25 - DEP:25 ARR:07 - DEP:07
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EDDS (STR) - Stuttgart
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
43
16. EDDT (TXL) - Berlin/ Tegel
Berlin Tegel has two parallel dependent runways operated in segregated mode, with very similar
results in terms of throughput and performance. The only impact is observed in the additional ASMA
times, higher for conf. ARR:08L-DEP:08R.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDT Berlin/ Tegel 191779 2 0.00 0.06 0.00 100.00%
71% 27%
24 24
29 2825 25
29 30
44 44
50 51
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0
10
20
30
40
50
60
ARR:26R - DEP:26L ARR:08L - DEP:08R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EDDT (TXL) - Berlin/ Tegel
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
44
17. EDDV (HAJ) - Hanover
Like Leipzig (EDDP), Hanover airport has not that many movements and despite having two parallel
runways, normally only one runway in mixed mode is operated (81% of the time). PSR shows only 1
or 2 movements difference for the most common configurations, and in fact it is lowest for the
segregated conf ARR:09L-DEP:09R (8% share). There is some imbalance in performance, especially
concerning the additional taxi-out times when using RWY 09L.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDV Hanover 64581 5 0.21 0.05 1.24 97.82%
39% 30% 9% 8% 3%
12 1113
79
17
15
13
8
10
12 12 12
109
17 1716
11
9
2122
20
16 16
26
29
22
1617
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
5
10
15
20
25
30
35
ARR:27R - DEP:27R ARR:27L - DEP:27L ARR:09R - DEP:09R ARR:09L - DEP:09R ARR:09L - DEP:09L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EDDV (HAJ) - Hanover
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
45
18. EDDW (BRE) - Bremen
Bremen operates only one runway (09/27) in both directions mixed mode, with identical result in
performance. In terms of PSR the impact is very low (only 1 movement difference) but the big
difference between the maximum observed throughput and the PSR signals the airport might be
underutilised which makes the PSR not a good proxy for capacity
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EDDW Bremen 29375 2 0.00 0.00 0.33 98.27%
67% 33%
10 10
19
1311 11
1715
1918
35
24
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0
5
10
15
20
25
30
35
40
ARR:27 - DEP:27 ARR:09 - DEP:09
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EDDW (BRE) - Bremen
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
46
19. EETN (TLL) - Tallinn
Tallinn is a one runway airport with a layout that clearly favours one runway direction utilization. In
fact, RWY26 is used (in mixed mode) 95% of the time in 2019. The other runway configuration is not
used more than 3% of the time (with some time intervals being the transition between configurations
or missing runway information) so it is not considered representative and there is no imbalance to be
analysed.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EETN Tallinn 43904 1 0.00 0.00 0.00 100.00%
95%
9
16
9
14 14
21
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0
5
10
15
20
25
ARR:26 - DEP:26
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EETN (TLL) - Tallinn
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
47
20. EFHK (HEL) - Helsinki/ Vantaa
Helsinki, with 3 runways (2 parallel plus one crossing) is operated in 6 representative configurations,
mostly in segregated mode. The results show significant differences in both throughput and
performance, where some configurations are clearly indicated for departure peaks and others for
arrival peaks. The 3 most common configurations are used in the departure peaks and result in lower
total throughput, higher additional taxi-out times and lower additional ASMA times. The two
configurations for the arrival peaks allow for higher total throughput but naturally increased additional
ASMA times. The most efficient configuration seems to be ARR:15-DEP22L (even if those are crossing
runways), when the number of departures and arrivals is balanced, but it is in use only 8% of the time.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EFHK Helsinki/ Vantaa 194634 6 1.00 0.26 6.77 95.84%
25% 24% 16% 8% 6% 5%
36 3733
37
1921
40 4139
41
22 22
29
22 21
38
4649
3633
24
40
5053
48 4844
5753
5760
58
50
62
57 58
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0
10
20
30
40
50
60
70
ARR:22L -DEP:22R
ARR:04L -DEP:04R
ARR:15 - DEP:22R ARR:15 - DEP:22L ARR:04L -ARR:04R -DEP:04R
ARR:22L -ARR:22R -DEP:22R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EFHK (HEL) - Helsinki/ Vantaa
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
48
21. EGBB (BHX) - Birmingham
Birmingham only has one runway that can be operated in both directions in mixed mode. The
performance for both runway directions use is very similar in throughput and additional taxi-out time,
and with higher additional ASMA for the most frequent configuration (RWY33). However, the analysis
also identifies as representative a configuration that implies opposite use of the runway: ARR:15-
DEP:33 (3% share). Normally this opposite use is observed of the periods of transition between runway
configurations, but the share is always much lower in that case. Another surprising aspect is that the
total throughput is almost the same for this configuration, and only additional taxi out time suffers a
considerable increase. This particular configuration would require further detailed analysis.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGBB Birmingham 107768 3 0.07 0.32 0.49 98.42%
49% 46% 3%
18 1819
2122 22
16 16 16
2021
20
2928 28
3635 35
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0
5
10
15
20
25
30
35
40
ARR:33 - DEP:33 ARR:15 - DEP:15 ARR:15 - DEP:33
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EGBB (BHX) - Birmingham
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
49
22. EGCC (MAN) - Manchester
Manchester, with 2 parallel dependent runways is operated in 4 configurations: 2 segregated (63%
total share) and 2 using only one runway in mixed mode (35% total share). The PSR is clearly lower for
the single runway use, and although performance is not much worse, maybe a segregated use in those
periods (if possible) could have reduced the delays.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGCC Manchester 202935 4 1.01 0.21 3.82 94.32%
50% 29% 13% 6%
30
24
29
22
35 36 35
2526
2225
20
3128 29
23
51
41
49
38
57
63
59
40
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0
10
20
30
40
50
60
70
ARR:23R - DEP:23L ARR:23R - DEP:23R ARR:05R - DEP:05L ARR:05L - DEP:05L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EGCC (MAN) - Manchester
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
50
23. EGGD (BRS) - Bristol
Bristol is a single runway airport, operated in mixed mode in both directions, being RWY27 the
preferential. Performance in terms of peak service rate differs in only 1 movement, and only additional
taxi-out times show worse performance for the non-preferential runway use.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGGD Bristol 66393 2 0.21 0.00 0.69 96.55%
69% 31%
12 12
1615
12 12
15 15
1920
2425
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0
5
10
15
20
25
30
ARR:27 - DEP:27 ARR:09 - DEP:09
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EGGD (BRS) - Bristol
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
51
24. EGGW (LTN) - London/ Luton
Luton has only one runway (08/26) that operates in both directions mixed mode, with identical results
in terms of PSR. The preferential runway (RWY26; 71% share) also has the lower additional times. In
general terms, no important imbalance is observed at this airport.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGGW London/ Luton 140958 2 0.13 0.16 0.00 100.00%
71% 29%
21 21
26 26
19 19
23 22
34 34
40 39
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0
5
10
15
20
25
30
35
40
45
ARR:26 - DEP:26 ARR:08 - DEP:08
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EGGW (LTN) - London/ Luton
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
52
25. EGKK (LGW) - London/ Gatwick
Gatwick is the busiest airport in Europe that is always operated with one runway in single mode. Given
the saturation, the PSR is a very good indication for capacity, and shows perfect symmetry in the
operation in terms of throughput. Additional taxi-out times however are impacted by the runway use,
being more than a minute higher for the most common runway configuration (RWY26L).
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGKK London/ Gatwick 284916 2 0.93 0.03 0.00 100.00%
69% 31%
33 34
39 38
28 28
3331
55 5558 59
0.0
2.0
4.0
6.0
8.0
10.0
12.0
0
10
20
30
40
50
60
70
ARR:26L - DEP:26L ARR:08R - DEP:08R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EGKK (LGW) - London/ Gatwick
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
53
26. EGLC (LCY) - London/ City
London City, with only one runway operated in both directions in mixed mode, shows nearly
symmetric PSR. Additional taxi-out time though, are considerably higher for most common runway in
use.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGLC London/ City 84208 2 0.93 0.00 0.31 99.03%
69% 31%
18 18
2221
17 17
2019
3231
3534
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0
5
10
15
20
25
30
35
40
ARR:27 - DEP:27 ARR:09 - DEP:09
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EGLC (LCY) - London/ City
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
54
27. EGLL (LHR) - London/ Heathrow
London Heathrow has two parallel runways operated always in segregated mode. The most common
runway direction is 27 alternating the arrivals and departures on 27R and 27L. While there is a perfect
symmetry in the PSR, performance results show very similar additional times, with additional taxi-out
times a little bit more impacted by the change of runway configuration.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGLL London/ Heathrow 478081 3 0.45 0.16 0.00 100.00%
36% 35% 25%
50 49 4953 54 54
45 44 4548 48 47
91 91 91
98 98 97
0.0
2.0
4.0
6.0
8.0
10.0
12.0
0
20
40
60
80
100
120
ARR:27L - DEP:27R ARR:27R - DEP:27L ARR:09L - DEP:09R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EGLL (LHR) - London/ Heathrow
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
55
28. EGNT (NCL) - Newcastle
Newcastle is a single runway airport, operated in mixed mode in both directions, being RWY25 the
preferential. Performance in terms of peak service rate differs in only 1 movement, and only additional
taxi-out times show worse performance for the non-preferential runway use. The big difference
between the maximum observed throughput and the PSR for the preferential runway use signals the
airport might be underutilised which makes the PSR not a good proxy for capacity
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGNT Newcastle 43438 2 0.08 0.03 0.72 95.74%
72% 27%
1011
17
14
10 10
20
1516
17
25
21
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0
5
10
15
20
25
30
ARR:25 - DEP:25 ARR:07 - DEP:07
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EGNT (NCL) - Newcastle
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
56
29. EGPD (ABZ) - Aberdeen
Aberdeen is a single runway airport (plus three helipads), operated in mixed mode in both directions,
with a very even use of both runway directions (55%-45%). Performance in terms of peak service rate
is identical for both configurations, but runway 34 shows longer additional times, mainly in the taxi-
out phase.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGPD Aberdeen 51233 2 0.23 0.13 0.00 100.00%
55% 45%
10 10
1312
10 10
15 15
17 17
2223
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
5
10
15
20
25
ARR:16 - DEP:16 ARR:34 - DEP:34
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EGPD (ABZ) - Aberdeen
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
57
30. EGPF (GLA) - Glasgow
Glasgow only has one runway that can be operated in both directions in mixed mode. The
performance for both runway directions use is similar in throughput (2 movements less for the
secondary runway use) showing higher taxi-out times for RWY 23 and higher additional ASMA times
for RWY 05. Nevertheless there is still a significant difference between the maximum throughput and
the PSR that might suggest the percentile 99 is not the best proxy for capacity.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGPF Glasgow 84345 2 0.16 0.04 0.54 97.86%
73% 27%
1615
22 22
14 14
1918
25
23
3130
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
0
5
10
15
20
25
30
35
ARR:23 - DEP:23 ARR:05 - DEP:05
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EGPF (GLA) - Glasgow
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
58
31. EGPH (EDI) - Edinburgh
Edinburgh is a single runway airport, operated in mixed mode in both directions, with almost identical
performance in terms of peak service rate. Additional times are slightly higher for the non-preferential
runway use.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGPH Edinburgh 131457 2 0.08 0.20 0.00 100.00%
68% 32%
1920
24 24
18 18
21 21
32 32
3635
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0
5
10
15
20
25
30
35
40
ARR:24 - DEP:24 ARR:06 - DEP:06
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EGPH (EDI) - Edinburgh
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
59
32. EGSS (STN) - London/ Stansted
Stansted is one of the top 30 airports in Europe, and it is a single runway airport. The runway is
operated in both directions in mixed mode, being RWY22 the most commonly used (71% share)
Performance in terms of throughput is completely symmetric, and additional times are very similar,
although the additional taxi out times are slightly higher for the preferential runway.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EGSS London/ Stansted 198511 2 0.33 0.04 0.00 100.00%
71% 29%
29 29
36 36
25 25
30 29
47 47
5351
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0
10
20
30
40
50
60
ARR:22 - DEP:22 ARR:04 - DEP:04
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EGSS (STN) - London/ Stansted
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
60
33. EHAM (AMS) - Amsterdam/ Schiphol
Amsterdam has 6 runways (more than any other airport in Europe) translating into many different
possible configurations. In addition, the 15 minutes intervals might not be long enough for all active
runways to be used in the interval. Nevertheless, the methodology detects all the 8 runway
configurations declared in the Airport Corner, and 4 additional ones. The usage of the configurations
is quite distributed, not having a clear preferential one. Although the PSR of total movements is quite
similar for all configurations, looking at the PSR for arrivals or departures it can be easily deducted
that the configurations are used either for arrival peaks or departure peaks. In terms of performance,
there are significant differences in both additional ASMA and additional taxi-out times amongst most
of the configurations, with configuration ARR:36C;ARR:36R-DEP:36L showing the worst results (more
than double compared to other configurations).
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EHAM Amsterdam/ Schiphol 509185 12 1.00 0.96 2.84 99.22%
12% 12% 8% 6% 6% 5% 4% 4% 4% 4% 4% 3%
40
78
40 40
79
42 42
75
46
77 7673
45
80
4441
82
44 43
76
46
78 77 7571
39
73 73
42
70 71
41
73
38 38 39
73
42
77 75
43
71 72
43
75
4138
43
102106 107 104
109104 104 106 107 108
103 103108
111 111 109112
107 107 106111 110
104 104
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
0
20
40
60
80
100
120
ARR:18C -ARR:18R -DEP:18L
ARR:18R -DEP:18L -
DEP:24
ARR:06 -ARR:36R -DEP:36L
ARR:18C -ARR:18R -
DEP:24
ARR:18R -DEP:18C -DEP:18L
ARR:36C -ARR:36R -DEP:36L
ARR:18R -ARR:22 -DEP:24
ARR:36R -DEP:36C -DEP:36L
ARR:27 -ARR:36C -DEP:36L
ARR:06 -DEP:36C -DEP:36L
ARR:27 -DEP:24 -DEP:36L
ARR:06 -DEP:09 -DEP:36L
Ad
dit
ion
al t
ime
s (m
inu
tes/
fligh
t)
Mo
vem
ents
/ho
ur
EHAM (AMS) - Amsterdam/ Schiphol
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
61
34. EICK (ORK) - Cork
Cork has two crossing runways but it only operates the longest one in mixed mode, with a very even
use of both runway directions (54%-45%). Performance in terms of peak service rate is almost identical
for both configurations, but RWY16 shows longer additional times, both in the approach and the taxi-
out phase.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EICK Cork 27002 2 0.10 0.26 0.52 97.90%
54% 45%
13 13
2018
13 13
1817
2425
3533
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0
5
10
15
20
25
30
35
40
ARR:16 - DEP:16 ARR:34 - DEP:34
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EICK (ORK) - Cork
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
62
35. EIDW (DUB) - Dublin
Dublin has two runways but operates only one of them in both directions mixed mode, with RWY28
as preferential (78% share). The PSR is almost identical for both configurations, but there is a clear
impact on performance when using RWY10, both in additional taxi-out times (bottleneck in the
departure queue for RWY10) and in ASMA times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EIDW Dublin 238044 2 0.42 0.47 0.20 99.59%
78% 20%
29 30
37 36
25 25
2927
49 48
5250
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0
10
20
30
40
50
60
ARR:28 - DEP:28 ARR:10 - DEP:10
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
EIDW (DUB) - Dublin
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
63
36. EKCH (CPH) - Copenhagen/ Kastrup
Copenhagen has 2 parallel runways and one crossing, but the crossing runway 12/30 is rarely used.
The two representative configurations use the parallel runways in segregated mode, with identical
total PSR. The preferential runway use ARR:22L-DEP:22R (73% share) shows slightly higher additional
times, both in the approach and the taxi-out phase.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EKCH Copenhagen/ Kastrup 263434 2 0.15 0.25 0.00 100.00%
73% 21%
36 37
43 43
3533
4239
61 61
6966
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
10
20
30
40
50
60
70
80
ARR:22L - DEP:22R ARR:04L - DEP:04R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EKCH (CPH) - Copenhagen/ Kastrup
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
64
37. ELLX (LUX) - Luxembourg
Luxembourg is a single runway airport, operated in mixed mode in both directions, being RWY24 the
most commonly used (68% share). Performance in terms of peak service rate differs in only 1
movement, and only additional taxi-out times show slightly worse performance for the preferential
runway use. The big difference between the maximum observed throughput and the PSR for the
preferential runway use signals the airport might be underutilised which makes the PSR not a good
proxy for capacity.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
ELLX Luxembourg 76300 2 0.20 0.01 0.68 97.50%
68% 32%
1516
2122
15 15
2019
2627
3634
0.0
0.5
1.0
1.5
2.0
2.5
0
5
10
15
20
25
30
35
40
ARR:24 - DEP:24 ARR:06 - DEP:06A
dd
itio
nal
tim
es (m
inu
tes/
fligh
t)
Mo
vem
ents
/ho
ur
ELLX (LUX) - Luxembourg
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
65
38. ENGM (OSL) - Oslo/ Gardermoen
Oslo has two parallel independent runways and operates them in many different ways, resulting in 8
configurations. About 42% of the time the runways are used in mixed mode independent, resulting in
the highest throughput and lower delays in the taxi-out and approach phases, but the other
configurations, using the runways as segregated, show lower PSR and especially much higher
additional taxi-out times (up to 7 minutes higher than other configurations for conf ARR:01R-DEP:01L)
This results in the highest impact on the additional taxi-out times of all 90 airports analysed. For about
9% of the time the airport also operates only RWY01L/19R (as single runway mixed mode), reducing
the PSR by more than 30 movements with respect to the reference capacity. Although this could be
due to a lack of demand, these configurations also observe worse additional taxi-out times than the
independent use of both runways.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
ENGM Oslo/ Gardermoen 251872 8 2.13 0.23 6.95 97.49%
23% 19% 18% 11% 8% 6% 5% 3%
40 4037 37
34
26
34
25
4744
41 43
3632
39
27
36 34 33 3430
22
3024
39 3835 35 34
26
3328
71 6965 64
57
41
55
38
7975 76
71
61
46
57
43
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0
10
20
30
40
50
60
70
80
90
ARR:01L -ARR:01R -DEP:01L -DEP:01R
ARR:19L -ARR:19R -DEP:19L -DEP:19R
ARR:01R -DEP:01L
ARR:19R -DEP:19L
ARR:19R -DEP:19L -DEP:19R
ARR:19R -DEP:19R
ARR:01L -DEP:01L -DEP:01R
ARR:01L -DEP:01L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
ENGM (OSL) - Oslo/ Gardermoen
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
66
39. EPWA (WAW) - Warszawa/ Chopina
Warsaw has 2 crossing runways operated in 5 different configurations. The high saturation level
ensures the PSR is a good proxy for capacity. The configuration that shows the best PSR (ARR:33-
DEP:29-DEP:33) is used only 3% of the time. In 2019, for about 40% of the time only RWY15/33 was in
use in mixed mode due to works on the other runway and although the impact on PSR was not very
high, the additional times were significantly longer, especially when using RWY15.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EPWA Warszawa/ Chopina 194160 5 0.70 0.61 3.40 97.42%
34% 29% 22% 11% 3%
27 27 28 2730
32 3134
3234
27 26 2724
26
32 3230
2628
42 41 4240
4546 47 47
42
48
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0
10
20
30
40
50
60
ARR:33 - DEP:29 ARR:33 - DEP:33 ARR:11 - DEP:15 ARR:15 - DEP:15 ARR:33 - DEP:29 -DEP:33
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EPWA (WAW) - Warszawa/ Chopina
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
67
40. ESGG (GOT) - Göteborg
Göteborg is a single runway airport, operated in mixed mode in both directions, being RWY21 the
most commonly used (70% share). Performance in terms of peak service rate differs in only 1
movement, and additional times show slightly worse performance for the non-preferential runway
use.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
ESGG Göteborg 69265 2 0.08 0.07 0.70 96.50%
70% 30%
12 12
17
15
12 12
16 16
1920
2524
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0
5
10
15
20
25
30
ARR:21 - DEP:21 ARR:03 - DEP:03
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
ESGG (GOT) - Göteborg
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
68
41. ESSA (ARN) - Stockholm/ Arlanda
Stockholm has three runways, two of them parallel. The analysis yields a maximum of two runways
are used at a time, always in segregated mode. Furthermore, for a 4% of the time only RWY01R is
used, in mixed mode, and this results in a significantly reduced PSR with consequent much higher
additional times. Even when using two runways there is a considerable lower PSR for configuration
ARR:26-DEP19L. Nevertheless, this configuration shows the lowest additional taxi-out times and
relatively low additional ASMA times, indicating that this configuration might be simply used in periods
with lower demand. In general the difference between the maximum throughput and the PSR signals
that congestion at Stockholm might not be enough to consider the PSR a good proxy for capacity.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
ESSA Stockholm/ Arlanda 233007 7 0.42 0.67 3.67 98.19%
24% 18% 18% 12% 9% 7% 4%
37 3835
32 32 3228
42 42 41
3537 37
3134 35
3230 30
32
24
38 3836
3235
37
25
6062
58
53
58 57
47
6770
6361
6361
48
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0
10
20
30
40
50
60
70
80
ARR:19L -DEP:19R
ARR:01R -DEP:01L
ARR:26 -DEP:19R
ARR:26 -DEP:19L
ARR:01L -DEP:08
ARR:19R -DEP:08
ARR:01R -DEP:01R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
ESSA (ARN) - Stockholm/ Arlanda
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
69
42. EVRA (RIX) - Riga
Riga is a single runway airport, operated in mixed mode in both directions, with almost identical
performance in terms of peak service rate. Additional times are slightly higher for the non-preferential
runway use.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EVRA Riga 86646 2 0.03 0.03 0.00 100.00%
.
63% 34%
20
18
27
24
17 17
21 21
24 24
30
28
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
5
10
15
20
25
30
35
ARR:18 - DEP:18 ARR:36 - DEP:36
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EVRA (RIX) - Riga
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
70
43. EYVI (VNO) - Vilnius
Vilnius, with only one runway operated in both directions in mixed mode, shows very similar PSR (only
one movement difference) and additional ASMA times. Additional taxi-out time though, are higher for
RWY01.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
EYVI Vilnius 46775 2 0.13 0.02 0.66 95.59%
66% 31%
8 8
16
109 9
1312
1415
1920
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0
5
10
15
20
25
ARR:19 - DEP:19 ARR:01 - DEP:01
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
EYVI (VNO) - Vilnius
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
71
44. GCFV (FUE) - Fuerteventura
Fuerteventura is a one runway airport with conditions that clearly favour one runway direction
utilization. In fact, RWY01 is used (in mixed mode) 97% of the time in 2019. The other runway
configuration is not used more than 3% of the time (with some time intervals being the transition
between configurations or missing runway information) so it is not considered representative and
there is no imbalance to be analysed.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
GCFV Fuerteventura 46356 1 0.00 0.00 0.00 100.00%
97%
12
16
12
16
21
28
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0
5
10
15
20
25
30
ARR:01 - DEP:01
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
GCFV (FUE) - Fuerteventura
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
72
45. GCLP (LPA) - Gran Canaria
Gran Canaria has two dependent parallel runways. For 56% of the time, only one of these two runways
are in use, in mixed mode, showing a much lower throughput than the combined use: conf ARR:03L-
DEP:03L;DEP:03R. This configuration shows the highest throughput but also the highest additional
taxi-out times, which might indicate is used for demand peaks.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
GCLP Gran Canaria 124152 3 0.34 0.16 6.77 84.15%
53% 40% 3%
17
23
14
20
27
1416
21
17
21
27
19
28
40
27
36
45
29
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
5
10
15
20
25
30
35
40
45
50
ARR:03L - DEP:03L ARR:03L - DEP:03L - DEP:03R ARR:03R - DEP:03R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
GCLP (LPA) - Gran Canaria
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
73
46. GCRR (ACE) - Lanzarote
Lanzarote is a one runway airport with conditions that clearly favour one runway direction utilization.
In fact, RWY03 is used (in mixed mode) 97% of the time in 2019. The other runway configuration is
not used more than 3% of the time (with some time intervals being the transition between
configurations or missing runway information) so it is not considered representative and there is no
imbalance to be analysed.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
GCRR Lanzarote 59130 1 0.00 0.00 0.00 100.00%
97%
12
18
13
16
22
26
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0
5
10
15
20
25
30
ARR:03 - DEP:03
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
GCRR (ACE) - Lanzarote
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
74
47. GCTS (TFS) - Tenerife Sur/Reina Sofia
Tenerife Sur has only one runway operated in both directions in mixed mode and conditions that
clearly favour one runway direction utilization (RWY07:95% share). The results shows very similar PSR
(only one movement difference) but additional times are higher for RWY25, although with low impact
due to the share (5%).
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
GCTS Tenerife Sur/Reina Sofia 68636 2 0.03 0.01 0.95 96.72%
95%5%
1617
20 20
1517
19 19
2829
3332
0.0
0.5
1.0
1.5
2.0
2.5
0
5
10
15
20
25
30
35
ARR:07 - DEP:07 ARR:25 - DEP:25
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
GCTS (TFS) - Tenerife Sur/Reina Sofia
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
75
48. GCXO (TFN) - Tenerife North
Tenerife North is a single runway airport operated in mixed mode in both directions, being RWY30 the
most commonly used (79% share). The PSR is identical for both configurations but additional times
are slightly higher for the non-preferential runway use.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
GCXO Tenerife North 72597 2 0.01 0.04 0.00 100.00%
79% 21%
13 13
1516
12 12
15 15
21 21
2625
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
0
5
10
15
20
25
30
ARR:30 - DEP:30 ARR:12 - DEP:12
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
GCXO (TFN) - Tenerife North
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
76
49. LBSF (SOF) - Sofia
Sofia operates only one runway (09/27) in both directions mixed mode, with almost identical result in
PSR. In terms of performance a slight increase of additional taxi-out times can be observed when using
RWY27 (64% share).
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LBSF Sofia 60266 2 0.12 0.01 0.00 100.00%
64% 35%
10 10
1413
1110
1514
17 17
23
21
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0
5
10
15
20
25
ARR:27 - DEP:27 ARR:09 - DEP:09
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LBSF (SOF) - Sofia
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
77
50. LCLK (LCA) - Larnaca
The data provided by Larnaca airport does not allow for the calculation of the performance indicators
(additional times) and the runway information is missing for 39% of the time, so the analysis in this
case cannot be performed.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LCLK Larnaca 60657 3 0.81 97.78%
44% 39% 14%
1112
10
13
21
1312 12
9
14
16
13
1920
17
2324
20
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0
5
10
15
20
25
30
ARR:22 - DEP:22 ARR:NA - DEP:NA ARR:04 - DEP:04
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LCLK (LCA) - Larnaca
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
78
51. LDZA (ZAG) - Zagreb
Zagreb is a single runway airport operated in mixed mode in both directions, being RWY05R the most
commonly used in 2019 (71% share). The PSR is identical for both configurations but additional times
are slightly higher for the non-preferential runway use.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LDZA Zagreb 44307 2 0.03 0.07 0.00 100.00%
71% 29%
10 10
13 13
9 9
1211
14 14
18 18
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0
2
4
6
8
10
12
14
16
18
20
ARR:05R - DEP:05R ARR:23L - DEP:23L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LDZA (ZAG) - Zagreb
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
79
52. LEAL (ALC) - Alicante
Alicante has one runway (10/28) that can only be operated in mixed mode in both directions. PSR is 2
movements higher for RWY10 (67% share) and performance in terms of additional taxi-out times is
very similar, Additional ASMA time when using RWY28 (33% share) are slightly higher.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LEAL Alicante 101210 2 0.01 0.13 0.65 97.89%
67% 33%
1817
23
2018
17
23
20
3129
37
34
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0
5
10
15
20
25
30
35
40
ARR:10 - DEP:10 ARR:28 - DEP:28
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LEAL (ALC) - Alicante
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
80
53. LEBB (BIO) - Bilbao
Bilbao has two runways but it normally uses only the longest one, operating it in mixed mode in both
directions. RWY30 is in use most of the time (81% share). Performance in terms of peak service rate
differs in only 1 movement, and RWY30 observed higher additional ASMA times while RWY12 shows
higher additional taxi-out times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LEBB Bilbao 48984 2 0.01 0.07 0.19 98.74%
81%19%
109
1312
9 9
1211
1514
1918
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0
2
4
6
8
10
12
14
16
18
20
ARR:30 - DEP:30 ARR:12 - DEP:12
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LEBB (BIO) - Bilbao
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
81
54. LEBL (BCN) - Barcelona
Barcelona has two parallel runways and one crossing. The crossing runway is normally used only
sometimes before 10h and after 19h, not reaching the 3% threshold of representative configurations
for the period analysed. The most common use is in segregated mode with a clear preference for
direction 25 (ARR:25R-DEP:25L; 81% share). This configuration shows a PSR one movement lower than
the opposite direction, but higher maximum throughput and lower additional times both in the
approach and in the taxi-out phase.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LEBL Barcelona 344508 2 0.10 0.15 0.81 98.87%
81% 17%
40 3944 42
38 38
44 42
71 72
7975
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0
10
20
30
40
50
60
70
80
90
ARR:25R - DEP:25L ARR:07L - DEP:07RA
dd
itio
nal
tim
es (m
inu
tes/
fligh
t)
Mo
vem
en
ts/h
ou
r
LEBL (BCN) - Barcelona
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
82
55. LEIB (IBZ) - Ibiza
Ibiza is a single runway airport, operated in mixed mode and using equally both runway directions
(51%-49%). Performance in terms of peak service rate differs in 2 movements, and although RWY06
shows the highest throughputs, it also shows higher additional ASMA and taxi-out times, which leads
to think that it is the configuration that was used when there was higher demand.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LEIB Ibiza 73356 2 0.29 0.15 1.02 96.71%
51% 49%
1617
22 22
1617
1921
2931
33
37
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
5
10
15
20
25
30
35
40
ARR:24 - DEP:24 ARR:06 - DEP:06
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LEIB (IBZ) - Ibiza
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
83
56. LEMD (MAD) - Madrid/ Barajas
Madrid has four runways, parallel two by two and operated in independent mode. The North
configuration is the preferred one (ARR:32L;ARR:32R-DEP36L;DEP36R; 69% share). The PSR differs
between configurations in one movement, but the significant difference between the maximum
throughput and the PSR signals the capacity might be higher than the percentile 99. In terms of
additional times, the best combined performance is found for the preferred configuration, having the
South configuration the worst additional taxi-out times and the configuration ARR:32L-
DEP36L;DEP36R almost triple the additional ASMA times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LEMD Madrid/ Barajas 426185 3 0.27 0.24 0.73 99.23%
69% 21% 4%
58 59 6167
7067
49 48
38
53 53
40
88 89 88
104100
93
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0
20
40
60
80
100
120
ARR:32L - ARR:32R - DEP:36L -DEP:36R
ARR:18L - ARR:18R - DEP:14L -DEP:14R
ARR:32L - DEP:36L - DEP:36RA
dd
itio
nal
tim
es (m
inu
tes/
fligh
t)
Mo
vem
en
ts/h
ou
r
LEMD (MAD) - Madrid/ Barajas
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
84
57. LEMG (AGP) - Málaga
Málaga has two non-parallel runways, but for 81% of the time the airport is operated as a single
runway in mixed mode using only RWY13/31. Being a seasonal airport is clear that for most of the
time the single runway use does not pose a problem, and although for these configurations the
throughput is much lower, it is probably just due to the lack of demand. For the segregated use
configurations, there are two (ARR:12-DEP:13 and ARR:31-DEP30) that have the highest peak service
rates (only one movement difference) but a significant imbalance in terms of additional times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LEMG Málaga 140721 5 0.93 0.42 10.86 85.10%
58% 23% 11% 3% 3%
2118
27
222526
22
30
24
28
2119
26
212425 25
28
2325
3532
47
38
4643
40
53
41
49
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0
10
20
30
40
50
60
ARR:13 - DEP:13 ARR:31 - DEP:31 ARR:12 - DEP:13 ARR:31 - DEP:30 -DEP:31
ARR:31 - DEP:30
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LEMG (AGP) - Málaga
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
85
58. LEPA (PMI) - Palma de Mallorca
Palma de Mallorca has two parallel runways most generally used in segregated mode, being direction
24 the preferred one (73% share). The PSR is almost identical for both configurations but performance
in terms of additional times is clearly better for configuration ARR:24L-DEP:24R, especially in the
approach. Being the most seasonal airport in the study, it would be interesting to analyse the PSR only
for the busiest months, which would probably be a better indication of the actual capacity.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LEPA Palma de Mallorca 217096 2 0.05 0.23 0.00 100.00%
73% 24%
35 34
4240
34 3539 38
65 65
73 72
0.0
0.5
1.0
1.5
2.0
2.5
0
10
20
30
40
50
60
70
80
ARR:24L - DEP:24R ARR:06L - DEP:06R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LEPA (PMI) - Palma de Mallorca
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
86
59. LEVC (VLC) - Valencia
Valencia is a single runway airport, operated in mixed mode and using almost equally both runway
directions (52%-48%). Performance in terms of peak service rate differs in one movement, and
although RWY12 shows the highest throughputs, it also shows higher additional ASMA and taxi-out
times, which leads to think that it is the configuration that was used when there was higher demand.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LEVC Valencia 72464 2 0.13 0.15 0.52 97.39%
52% 48%
12 12
15
17
1112
1415
1920
23
26
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0
5
10
15
20
25
30
ARR:30 - DEP:30 ARR:12 - DEP:12
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LEVC (VLC) - Valencia
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
87
60. LEZL (SVQ) - Sevilla
Sevilla is a single runway airport, operated in mixed mode in both directions. The peak service rate for
both runway directions differs in one movement, but performance in terms of additional times is
nearly identical, showing no real imbalance.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LEZL Sevilla 58721 2 0.02 0.01 0.62 96.73%
62% 38%
1112
1615
11 11
1615
1819
2423
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0
5
10
15
20
25
30
ARR:27 - DEP:27 ARR:09 - DEP:09
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LEZL (SVQ) - Sevilla
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
88
61. LFBO (TLS) - Toulouse-Blagnac
Toulouse airport has two parallel runways (plus one helipad) operated in several different ways, from
segregated to single runway use in mixed mode, being the share quite distributed in 7 configurations.
There are several movements difference in throughput between configurations and the highest PSR
value is observed for conf ARR:14R-DEP:14L (29% share).The data provision from Toulouse does not
allow the analysis of the additional times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LFBO Toulouse-Blagnac 95665 7 1.76 97.29%
35% 29% 9% 7% 6% 6% 5%
15 1513 13 13
11
15
1819
14 1413
1416
1314
1315
1213
1515
19
1416
1415
17
2426
22
25
21 21
25
3029
23
27
2223
28
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0
5
10
15
20
25
30
35
ARR:32L -DEP:32R
ARR:14R -DEP:14L
ARR:32R -DEP:32R
ARR:32L -ARR:32R -DEP:32R
ARR:14R -DEP:14R
ARR:32L -DEP:32L
ARR:14L -DEP:14L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LFBO (TLS) - Toulouse-Blagnac
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
89
62. LFLL (LYS) - Lyon-Saint-Exupéry
Lyon has two parallel runways but it normally operates only the longest one in mixed mode, in either
direction. The PSR only differs in one movement, and performance in terms of additional times is very
similar, with only slightly higher additional ASMA times for RWY 35L (59% share). The usage of only
one runway plus the deviation between PSR and maximum throughput signal that congestion might
be too low to consider the PSR a good proxy for capacity.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LFLL Lyon-Saint-Exupéry 116451 2 0.01 0.09 0.39 98.95%
59% 39%
25 25
31 31
26 25
3331
37 36
45 45
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0
5
10
15
20
25
30
35
40
45
50
ARR:35L - DEP:35L ARR:17R - DEP:17R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LFLL (LYS) - Lyon-Saint-Exupéry
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
90
63. LFML (MRS) - Marseille-Provence
Marseille has two parallel runways but like Lyon it normally operates only the longest one in mixed
mode, in both directions. Performance in terms of PSR is identical with slightly higher additional ASMA
times for RWY 31R (66% share). The data provision from Marseille does not allow for the analysis of
the additional taxi-out times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LFML Marseille-Provence 102011 2 0.15 0.00 100.00%
66% 32%
17 17
21 21
16 16
2019
27 27
3231
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0
5
10
15
20
25
30
35
ARR:31R - DEP:31R ARR:13L - DEP:13L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LFML (MRS) - Marseille-Provence
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
91
64. LFMN (NCE) - Nice-Côte d’Azur
Nice airport has two parallel runways operated in several different ways, from segregated to single
runway use in mixed mode, with very different results in throughput. However it is important to
consider the high seasonality of Nice, and in fact the most common single runway use (RWY04R; 29%
share) shows the lowest PSR but a significantly higher maximum throughput and the best performance
in terms of additional times, so it is probably a configuration used in low demand. The two
configurations using the direction 22 show much higher additional times in the approach.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LFMN Nice-Côte d’Azur 145645 5 0.33 0.52 8.99 89.15%
51% 29% 8% 6% 4%
26
18 19
27 27
33
2523
37
32
25
18 18
2427
29
2321
2731
47
3235
4851
56
4339
6157
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0
10
20
30
40
50
60
70
ARR:04L - DEP:04R ARR:04R - DEP:04R ARR:22L - DEP:22L ARR:22R - DEP:22L ARR:04L - ARR:04R -DEP:04R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LFMN (NCE) - Nice-Côte d’Azur
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
92
65. LFPG (CDG) - Paris-Charles-de-Gaulle
Paris Charles de Gaulle has 4 runways, almost parallel that are used as two independent sets of two
runways operating in segregated mode. The longest runways are always used for departures and the
shortest for arrivals. The capacity is nearly symmetric according to the results and the performance in
terms of additional times is also very similar for both configurations.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LFPG Paris-Charles-de-Gaulle 504887 2 0.08 0.03 0.62 99.43%
62% 35%
66 66
75 74
61 6268 68
108 109116 115
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0
20
40
60
80
100
120
140
ARR:26L - ARR:27R - DEP:26R - DEP:27L ARR:08R - ARR:09L - DEP:08L - DEP:09R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LFPG (CDG) - Paris-Charles-de-Gaulle
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
93
66. LFPO (ORY) - Paris-Orly
Paris Orly has three runways, from which RWY02/20 is operated only in exceptional circumstances.
During the second part of the year the runway 08/26 was completely rebuilt, deeply affecting the
normal operating conditions of the airport and consequently the results of this study. Taking the entire
year into account, the most common use is the segregated modes ARR:26-DEP:24 (41% share) and
ARR:06-DEP:08 (23% share), but of course due to the runway closure the single runway use in mixed
mode also appears as representative. Although the schedule was adapted and reductions were
coordinated during the period of the works, the impact of the single runway use and the works is clear
in terms of additional taxi-out and ASMA times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LFPO Paris-Orly 221602 4 0.54 0.25 3.18 97.41%
41% 23% 18% 12%
29 3026 26
36
32 31 302931
26 26
35 36
28 28
5254
46 46
62 62
5048
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0
10
20
30
40
50
60
70
ARR:26 - DEP:24 ARR:06 - DEP:08 ARR:24 - DEP:24 ARR:06 - DEP:06
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LFPO (ORY) - Paris-Orly
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
94
67. LFSB (BSL) - Bâle-Mulhouse
Bâle-Mulhouse has two crossing runways but one is too short so it normally uses only the longest one,
operating it in mixed mode in both directions. RWY15 is used 91% of the time in 2019, showing a PSR
1 movement higher than the opposite direction, but this difference is minimized by the share and the
resilience remains high. Nevertheless, the additional ASMA times are higher for this configuration,
while additional taxi-out times are lower.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LFSB Bâle-Mulhouse 84560 2 0.04 0.52 0.09 99.63%
91% 9%
14 14
28
18
13 13
19
15
2423
36
27
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0
5
10
15
20
25
30
35
40
ARR:15 - DEP:15 ARR:33 - DEP:33
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LFSB (BSL) - Bâle-Mulhouse
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
95
68. LGAV (ATH) - Athens
Athens has two parallel runways operated in segregated mode, with three representative runway
configurations. The most common configuration ARR:03L-DEP:03R (50% share) shows the highest
throughput (both in PSR and in maximum) but it also has the highest additional taxi-out times, while
showing the best performance in the approach. There is a four movements difference between the
two most common configurations, affecting the resilience and the impact on the PSR.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LGAV Athens 220639 3 0.33 0.14 1.15 97.87%
50% 29% 14%
3229
31
3734 34
3027 28
3431 32
54
50
54
6057 58
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
10
20
30
40
50
60
70
ARR:03L - DEP:03R ARR:21R - DEP:21L ARR:03R - DEP:03L
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LGAV (ATH) - Athens
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
96
69. LHBP (BUD) - Budapest/ Ferihegy
Budapest airport has two parallel runways, only used simultaneously in segregated mode 65% of the
time (the rest of the time only one runway in mixed mode is in use). The PSR is only two movements
lower for the single runway configurations than for the segregated mode, but the performance in
terms of additional times is much worse for both taxi-out and approach phase when only one runway
is used for both arrivals and departures.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LHBP Budapest/ Ferihegy 122132 5 0.76 0.45 0.57 99.00%
43% 22% 16% 10% 6%
18 18 18 18 17
2122 22
2022
1918 18 17 18
23 23 23
20 21
32 3230 30 31
3937
3234
37
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0
5
10
15
20
25
30
35
40
45
ARR:31R - DEP:31L ARR:13R - DEP:13L ARR:31L - DEP:31L ARR:13R - DEP:13R ARR:31R - DEP:31R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LHBP (BUD) - Budapest/ Ferihegy
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
97
70. LICC (CTA) - Catania
Catania is a one runway airport with conditions that clearly favour one runway direction utilization. In
fact, RWY01 is used (in mixed mode) 97% of the time in 2019. The other runway configuration is not
used more than 3% of the time (with some time intervals being the transition between configurations
or missing runway information) so it is not considered representative and there is no imbalance to be
analysed.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LICC Catania 75399 1 0.00 0.00 0.00 100.00%
100%
12
16
12
16
21
27
2.2
2.3
2.3
2.3
2.3
2.3
2.4
2.4
2.4
2.4
0
5
10
15
20
25
30
ARR:08 - DEP:08
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LICC (CTA) - Catania
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
98
71. LIMC (MXP) - Milan/ Malpensa
Milan Malpensa has two parallel runways operated in segregated mode and almost equally used in
both directions. The PSR is 2 movements higher for the more common configuration ARR:35R-
DEP:35L; 51% share) and this configuration also shows slightly better performance in terms of
additional times, especially in the approach.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LIMC Milan/ Malpensa 233978 2 0.01 0.25 0.90 98.55%
51% 45%
3533
39 39
32 31
3734
6260
69 68
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0
10
20
30
40
50
60
70
80
ARR:35R - DEP:35L ARR:35L - DEP:35R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LIMC (MXP) - Milan/ Malpensa
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
99
72. LIME (BGY) - Bergamo
Bergamo has two runways but one is a short landing strip, so basically it is operated as a one runway
airport. According to the reported runway use, RWY28 is used (in mixed mode) a 100% of the time in
2019. There is therefore no imbalance to be analysed. During the first half of 2019 the systems at
Bergamo could not register properly the movements on RWY10, and RWY28 was the default value.
Nevertheless, in the second half of the year, although some movements are observed on RWY10, the
share is minimal (less than 1%).
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LIME Bergamo 95147 1 0.00 0.00 0.00 100.00%
100%
14
19
14
20
23
30
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
0
5
10
15
20
25
30
35
ARR:28 - DEP:28A
dd
itio
nal
tim
es (m
inu
tes/
fligh
t)
Mo
vem
ents
/ho
ur
LIME (BGY) - Bergamo
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
100
73. LIMF (TRN) - Torino Caselle
Torino is a one runway airport with conditions that clearly favour one runway direction utilization. In
fact, RWY36 is used (in mixed mode) a 100% of the time in 2019. There is therefore no imbalance to
be analysed.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LIMF Torino Caselle 39229 1 0.00 0.00 0.00 100.00%
100%
9
15
9
17
15
23
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0
5
10
15
20
25
ARR:36 - DEP:36
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LIMF (TRN) - Torino Caselle
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
101
74. LIML (LIN) - Milan/ Linate
Milan Linate operates as a one runway airport and its conditions clearly favour one runway direction
utilization. In fact, RWY36 is used (in mixed mode) a 100% of the time in 2019. There is therefore no
imbalance to be analysed.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LIML Milan/ Linate 84458 1 0.00 0.00 0.00 100.00%
100%
17
24
16
21
29
36
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
5
10
15
20
25
30
35
40
ARR:36 - DEP:36
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LIML (LIN) - Milan/ Linate
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
102
75. LIPE (BLQ) - Bologna
Bologna is a single runway airport operated in mixed mode in both directions, being RWY12 the most
commonly used in 2019 (76% share). The PSR is almost identical for both configurations but additional
times are slightly higher for the preferential runway use.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LIPE Bologna 77090 2 0.16 0.05 0.00 100.00%
76% 23%
1112
1415
12 12
1516
19 19
2221
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0
5
10
15
20
25
ARR:12 - DEP:12 ARR:30 - DEP:30
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LIPE (BLQ) - Bologna
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
103
76. LIPZ (VCE) - Venice
Venice is has two runways but one is used mainly as taxiway. Conditions clearly favour one runway
direction utilization, resulting in one configuration, where only RWY04R is used (in mixed mode) a
100% of the time in 2019. There is therefore no imbalance to be analysed.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LIPZ Venice 95266 1 0.00 0.00 0.00 100.00%
100%
16
20
15
20
27
34
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
5
10
15
20
25
30
35
40
ARR:04R - DEP:04R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LIPZ (VCE) - Venice
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
104
77. LIRA (CIA) - Rome/Ciampino
Rome Ciampino is a single runway airport operated in mixed mode in both directions, with RWY15 in
use most of the time (91% share). The PSR is identical for both configurations but additional ASMA
times are slightly higher for the non-preferential runway use.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LIRA Rome/Ciampino 51154 2 0.00 0.03 0.00 100.00%
91%8%
109
14
1110
9
16
11
16 16
22
18
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0
5
10
15
20
25
ARR:15 - DEP:15 ARR:33 - DEP:33
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LIRA (CIA) - Rome/Ciampino
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
105
78. LIRF (FCO) - Rome/Fiumicino
Rome Fiumicino has three runways, two of them parallel, and from all operating possibilities the
analysis identifies three representative configurations. The PSR is two movements lower for the least
common configuration, with a low impact on resilience. The additional times at Rome, especially on
the taxi-out phase are very high in general. There is no clear configuration that performs better than
the others for both indicators, one being better for taxi and a different one being better for the
approach.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LIRF Rome/Fiumicino 309783 3 0.12 0.36 0.24 99.68%
60% 14% 12%
37
45
36
4549
40
49
34
48
56
39
52
74 74 72
8581
76
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
0
10
20
30
40
50
60
70
80
90
ARR:16L - ARR:16R - DEP:25 ARR:16L - DEP:25 ARR:34L - ARR:34R - DEP:25A
dd
itio
nal
tim
es (m
inu
tes/
fligh
t)
Mo
vem
ents
/ho
ur
LIRF (FCO) - Rome/Fiumicino
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
106
79. LIRN (NAP) - Naples
Naples has one runway operated in mixed mode with quite even share of both directions (58%-40%).
Performance in terms of peak service rate differs in two movements, and while RWY24 observed
higher additional taxi-out times, RWY06 shows significantly higher additional ASMA times
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LIRN Naples 84387 2 0.16 0.32 1.15 95.73%
58% 40%
1415
17
19
1415
1819
25
27
32 32
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
5
10
15
20
25
30
35
ARR:24 - DEP:24 ARR:06 - DEP:06
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LIRN (NAP) - Naples
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
107
80. LJLJ (LJU) - Ljubljana
Ljubljana is a one runway airport with conditions that clearly favour one runway direction utilization.
In fact, RWY30 is used (in mixed mode) 98% of the time in 2019. The other runway configuration is
not considered representative and there is no imbalance to be analysed.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LJLJ Ljubljana 27935 1 0.00 0.00 0.00 100.00%
98%
9
13
10
1314
18
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
0
2
4
6
8
10
12
14
16
18
20
ARR:30 - DEP:30
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LJLJ (LJU) - Ljubljana
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
108
81. LKPR (PRG) - Prague
Prague has two crossing runways but it normally only operates one of them at a time, in mixed mode.
The PSR only differs in one movement less for the least common runway (RWY30; 5% share), resulting
in high resilience. Performance in terms of additional times is best for both approach and taxi-out
phase for the RWY06 (24% share) and not for the preferred runway (RWY24; 69% share).
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LKPR Prague 150434 3 0.56 0.18 0.05 99.87%
69% 24% 5%
23 23 24
3028 28
24 24 23
2931
26
39 39 38
47
4442
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0
5
10
15
20
25
30
35
40
45
50
ARR:24 - DEP:24 ARR:06 - DEP:06 ARR:30 - DEP:30
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LKPR (PRG) - Prague
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
109
82. LMML (MLA) - Malta
Malta airport has two converging runways but it normally uses only the longest one, operating it in
mixed mode in both directions. The most commonly used (RWY31; 59%) shows a PSR significantly
higher than the opposite direction, which results in a lower resilience value. Nevertheless, the
additional taxi-out times are also higher for this configuration, while additional ASMA times are very
similar.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LMML Malta 58181 2 0.17 0.03 1.59 91.63%
59% 32%
11
9
15
1211
9
1413
19
14
23
20
0.0
0.5
1.0
1.5
2.0
2.5
0
5
10
15
20
25
ARR:31 - DEP:31 ARR:13 - DEP:13
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LMML (MLA) - Malta
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
110
83. LOWW (VIE) - Vienna
Vienna airport has two converging runways that are operated in 7 representative configurations
according to the data found in 2019. The PSR for all of the configurations that use the two runways
differs in one or two movements, but in addition there is a configuration in which only RWY29 is in
use (5% share) with a significant reduction of the PSR but most importantly, a drastic increase of the
additional times both in the taxi-out and the approach phase. It is also noticeable the configurations
used in departure or arrival peaks, which show the highest throughputs.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LOWW Vienna 281716 7 0.53 1.29 2.71 98.14%
40% 16% 14% 12% 7% 5% 4%
37 37
4541
2833
4443 43
52
45
3336
50
40 39
28
37
44
2927
43 42
31
41
47
31 32
62 6265 63 64
50
65
7268
7267 67
57
76
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
0
10
20
30
40
50
60
70
80
ARR:34 -DEP:29
ARR:16 -DEP:29
ARR:34 -DEP:29 -DEP:34
ARR:11 -DEP:16
ARR:11 -ARR:16 -DEP:16
ARR:29 -DEP:29
ARR:16 -DEP:16 -DEP:29
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LOWW (VIE) - Vienna
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
111
84. LPFR (FAO) - Faro
Faro is a single runway airport operated in mixed mode in both directions, being RWY28 the most
commonly used in 2019 (73% share). The PSR for both configurations differs only in one movement
and additional times are very similar, with additional ASMA time slightly higher for the preferential
runway use.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LPFR Faro 60664 2 0.03 0.09 0.73 97.30%
73% 27%
15 15
2021
15 16
20 20
2627
3335
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0
5
10
15
20
25
30
35
40
ARR:28 - DEP:28 ARR:10 - DEP:10
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LPFR (FAO) - Faro
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
112
85. LPPR (OPO) - Porto
Porto in a single runway airport, used in mixed mode in both directions.RWY35, with a 69% share,
shows the best performance both in throughput and in terms of additional times both in the taxi-out
and approach phase, which are significantly worse for RWY17 (31% share).
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LPPR Porto 98816 2 0.50 0.52 0.31 98.70%
69% 31%
14 14
17 17
13 13
18
16
2423
30
27
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0
5
10
15
20
25
30
35
ARR:35 - DEP:35 ARR:17 - DEP:17
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LPPR (OPO) - Porto
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
113
86. LPPT (LIS) - Lisbon
Lisbon airport has one runway (the former runway 17/35 was reconverted into a taxiway) used in
mixed mode in both directions. The difference in PSR is only one movement less for RWY 21, and
although this configuration shows higher additional taxi-out times, the most common runway
(RWY03; 72% share) shows higher additional ASMA times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LPPT Lisbon 220938 2 0.16 0.30 0.28 99.35%
72% 28%
24 23
29
26
23 22
25 24
43 42
45 45
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0
5
10
15
20
25
30
35
40
45
50
ARR:03 - DEP:03 ARR:21 - DEP:21
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LPPT (LIS) - Lisbon
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
114
87. LROP (OTP) - Bucharest/ Otopeni
Bucharest Otopeni has two parallel runways used in segregated mode for 82% of the time. There is up
to four movements difference in PSR between configurations, but it is interesting to observe that the
use of only RWY08L in mixed mode is the second highest in terms of PSR. The single runway
configurations (either RWY08L or RWY08R) show however the longest additional taxi-out times, while
additional ASMA times are higher for two of the segregated configurations.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LROP Bucharest/ Otopeni 122831 5 0.28 0.28 1.98 96.91%
36% 24% 22% 7% 5%
22
1819
1816
29
2325
23 23
1816
1718 18
2120
21 2120
31
2728
29
26
36
33
37
3331
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0
5
10
15
20
25
30
35
40
ARR:08L - DEP:08R ARR:08R - DEP:08L ARR:26L - DEP:26R ARR:08L - DEP:08L ARR:08R - DEP:08R
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
ents
/ho
ur
LROP (OTP) - Bucharest/ Otopeni
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
115
88. LSGG (GVA) - Geneva
Geneva is a single runway airport operated in mixed mode in both directions. RWY22, used 59% of the
time, has a PSR 1 one movement lower than RWY04, and additional times are higher both in the
approach and the taxi-out phase.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LSGG Geneva 179115 2 0.20 0.34 0.59 98.67%
59% 41%
25 25
30 30
24 2427 28
43 4446
48
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0
10
20
30
40
50
60
ARR:22 - DEP:22 ARR:04 - DEP:04
Ad
dit
ion
al ti
mes
(min
ute
s/fli
ght)
Mo
vem
en
ts/h
ou
r
LSGG (GVA) - Geneva
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
116
89. LSZH (ZRH) - Zürich
Zürich has three runways, operated most of the time in segregated mode. The most common
configuration ARR:14-DEP:28 (42% share) has the highest throughput and the lowest combined
delays. The least common configuration ARR:34-DEP:32 (5% share) shows a PSR 10 movements below
the reference capacity, and the highest delays in the taxi-out and approach while configuration
ARR:14-DEP:16;DEP:28 (30% share) seems to be used for departure peaks and has high additional taxi-
out times.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LSZH Zürich 269223 4 1.19 0.54 2.06 98.50%
42% 30% 19% 5%
36
41
3431
3943
3835
40
33 3330
43
36 36
31
6462
59
54
7268
63
56
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0
10
20
30
40
50
60
70
80
ARR:14 - DEP:28 ARR:14 - DEP:16 - DEP:28 ARR:28 - DEP:32 ARR:34 - DEP:32A
dd
itio
nal
tim
es (m
inu
tes/
fligh
t)
Mo
vem
ents
/ho
ur
LSZH (ZRH) - Zürich
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL
117
90. LZIB (BTS) - Bratislava
Bratislava has two crossing runways operated in four representative configurations. The configuration
that shows the highest throughput and the best performance in terms of delay in the taxi-out phase
is used only 5% of the time (ARR:22-DEP:13). Although the demand is probably low to consider the
PSR a good proxy for capacity, the two most common configurations, with a combined share of 89%
show a low departure throughput with much higher additional taxi-out times, which might signal
issues like bottlenecks in the taxi-out flows.
Airport ICAO Airport Name
Annual traffic 2019
Number of configura-tions
Impact Add. Taxi-Out Time (min/dep)
Impact Add. ASMA Time (min/arr)
Impact Peak Service Rate (mov/h)
Airport Resilience
LZIB Bratislava 22593 4 0.51 0.07 6.22 72.60%
71% 18% 5% 4%
7 7
9
7
1110
13
9
7 7
10
6
109
12
7
1112
18
11
1415
23
12
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0
5
10
15
20
25
ARR:31 - DEP:04 ARR:31 - DEP:31 ARR:22 - DEP:13 ARR:31 - DEP:13A
dd
itio
nal
tim
es (m
inu
tes/
fligh
t)
Mo
vem
en
ts/h
ou
r
LZIB (BTS) - Bratislava
Add TXOT Add ASMA P99 DEP MAX DEP P99 ARR MAX ARR P99 TOTAL MAX TOTAL