Procedure to generate synthetic data for computational experiments on the single airport slot allocation problem.
Inspired by slot request data sets published by Agencia Nacional de Aviacão Civil (ANAC) we create distributions for a set of parameters that characterise airport demand:
- total number of operations in a summer season
- domestic/international market split
- proportion of flight series including turnaround information
- distribution of requested turnaround times
- distribution of seats per aircraft
- distribution of seat load factors
- distribution of start and end dates of the series
- distribution of weekly frequencies
Inspired by capacity declaration reports published by Airport Coordination Limited (ACL), we create different sets of runway / terminal capacity limits characterised by:
- Type of resource (runway / terminal)
- Arrival / departure / all flights
- Domestic / international / all flights
- Duration of time window each individual capacity limit
- Frequency of start times of each individual capacity limit
- Level of congestion
To generate a single instance, we sample from demand distributions to generate a slot request dataset and we choose a set of capacity constraints and determine the declared limits based on the generated demand to produce a capacity data set.
Possible distributions for each parameter can be found in parameters.yml.
This project used Python 3.9. Install dependencies with
$ pip install -r requirements.txt
Generate synthetic instances using
$ python src/generate.py [NUMBER_OF_INSTANCES]
This will create [NUMBER_OF_INSTANCES] pairs of capacity and demand files. Demand files will be stored in schedules/demand/, with the name IXXXX_demand.csv, where XXXX will be a unique identifier of the instance. Capacity files will be stored in schedules/capacity/ with the name IXXXX_capacity.csv, where XXXX will math the unique identifier of its corresponding demand file. It will also create a file IXXXX_metadata.yml for each instance in schedules/metadata/ which will show the distributions and parameters selected for that instance.
Generate PDF reports showing summary statistics of each generated instance found in the schedules folder using
$ python src/visualise.py
Clean all generated instances, metadata and reports using
$ python src/clean.py
Fields in demand file IXXXX_demand.csv:
| Field | Description |
|---|---|
FREQ |
This field indicates days of the week that a series requests to operate. e.g. "01034007" indicates that the series requests to operate on Tuesdays (1), Thursdays (3), Fridays (4) and Sundays (7) |
Carrier |
3-letter code of dummy name of carrier operating the flight in the series |
FlNum |
Flight number of flights in the series |
Airport |
3-letter code of dummy airport associated with this instance |
Season |
3-letter code in format XYY indicating aviation season under consideration, where "X" indicates whether it's a winter (W) or summer (S) season and "YY" is the year in 2-digit format. All instances generated are based on the 2020 summer season (S20) |
ServType |
Type of service, "J" for scheduled in all generated instances |
Term |
Name of terminal in which each series operates, e.g. Term1, Term2 |
OrigDest |
3-letter code of dummy origin or destination airport. "ZZD" denotes a domestic airport and "ZZI" an international airport |
StartDate |
First date in the season that the series requests to operate |
EndDate |
Last date in the season that the series requests to operate |
Seats |
Number of seats in every flight in the series |
Pax |
Number of passengers in every flight in the series |
ArrDep |
Arrival or departure |
Req |
Requested slot, in HHMM format |
NoOps |
Number of requested operations in the season. It can be calculated from StartDate, EndDate and FREQ |
TurnCarrier |
Name of the carrier operating the turnaround flight, if this information is provided |
TurnFlNum |
Flight number of the turnaround flight, if this information is provided |
Fields in demand file IXXXX_capacity.csv:
| Field | Description |
|---|---|
Constraint |
Unique identifier of a type of capacity constraint, which groups all individual limits with the same Resource, ArrDep, Duration, DomInt and Terminal |
Resource |
Runway (limits the number of flights) or terminal (number of passengers) |
ArrDep |
Whether this limit affects arriving flights (A), departures (D) or all flights (T) |
Duration |
Duration of the time window of the capacity limit, in minutes |
Limit |
Maximum number of flights or passengers allowed within the time window |
Time |
Start time of the capacity limit window, in HHMM format |
DomInt |
Whether this limit affects domestic flights (D), international flights (I) or all flights (T) |
Terminal |
Capacity limit only affects flights operating in this terminal. If empty, flights in any terminal can be affected |
This code was used to generate 100 synthetic instances uploaded to DataShare Edinburgh. Below you can find two tables with summary statistics for each of the instances contained in that data repository.
In the demand table, Series represents the number of series in each instance; Movements is the total number of requested operations throughout the season; Total displacement (min) is the total displacement in minutes across all operations. There is no column for rejections because none of these instances resulted in any rejections.
| Instance | Series | Movements | Total displacement (min) |
|---|---|---|---|
| I0000 | 1,920 | 84,075 | 22,365 |
| I0001 | 3,041 | 134,068 | 70,310 |
| I0002 | 3,942 | 183,014 | 70,310 |
| I0003 | 3,126 | 137,050 | 1,265 |
| I0004 | 2,303 | 197,047 | 26,490 |
| I0005 | 1,333 | 58,031 | 51,105 |
| I0006 | 2,801 | 99,104 | 10,150 |
| I0007 | 1,538 | 136,052 | 196,200 |
| I0008 | 1,032 | 92,069 | 6,430 |
| I0009 | 1,062 | 62,029 | 21,260 |
| I0010 | 805 | 37,058 | 9,035 |
| I0011 | 4,863 | 161,053 | 30,735 |
| I0012 | 2,070 | 136,050 | 289,025 |
| I0013 | 2,151 | 190,071 | 30,800 |
| I0014 | 3,595 | 129,002 | 16,160 |
| I0015 | 1,381 | 81,002 | 17,240 |
| I0016 | 1,612 | 72,042 | 43,000 |
| I0017 | 1,396 | 121,014 | 108,395 |
| I0018 | 1,320 | 84,358 | 18,415 |
| I0019 | 1,210 | 59,049 | 20,665 |
| I0020 | 1,054 | 47,000 | 25,735 |
| I0021 | 343 | 22,375 | 63,835 |
| I0022 | 2,777 | 128,081 | 44,275 |
| I0023 | 1,382 | 97,107 | 17,825 |
| I0024 | 2,592 | 116,006 | 62,145 |
| I0025 | 938 | 53,276 | 36,785 |
| I0026 | 4,100 | 178,419 | 115,060 |
| I0027 | 3,559 | 199,069 | 212,485 |
| I0028 | 2,234 | 187,015 | 41,555 |
| I0029 | 117 | 10,278 | 1,575 |
| I0030 | 1,401 | 96,017 | 16,630 |
| I0031 | 1,463 | 128,158 | 77,110 |
| I0032 | 727 | 33,329 | 18,215 |
| I0033 | 1,393 | 122,379 | 23,980 |
| I0034 | 337 | 30,161 | 17,385 |
| I0035 | 1,874 | 83,051 | 7,085 |
| I0036 | 1,551 | 55,051 | 217,315 |
| I0037 | 1,128 | 99,310 | 17,825 |
| I0038 | 2,692 | 162,118 | 12,430 |
| I0039 | 795 | 55,051 | 129,510 |
| I0040 | 1,384 | 57,206 | 19,190 |
| I0041 | 2,245 | 152,182 | 94,760 |
| I0042 | 2,142 | 146,265 | 18,965 |
| I0043 | 4,575 | 199,049 | 46,255 |
| I0044 | 3,197 | 139,076 | 30,990 |
| I0045 | 286 | 23,025 | 5,775 |
| I0046 | 234 | 23,137 | 2,330 |
| I0047 | 2,165 | 143,016 | 87,970 |
| I0048 | 1,953 | 168,015 | 517,755 |
| I0049 | 1,933 | 131,019 | 16,625 |
| I0050 | 501 | 21,111 | 19,205 |
| I0051 | 865 | 36,020 | 19,435 |
| I0052 | 2,778 | 188,118 | 162,160 |
| I0053 | 958 | 81,023 | 30,295 |
| I0054 | 328 | 19,002 | 8,595 |
| I0055 | 1,513 | 127,022 | 31,150 |
| I0056 | 562 | 20,054 | 12,130 |
| I0057 | 2,051 | 145,134 | 22,690 |
| I0058 | 4,481 | 156,043 | 133,855 |
| I0059 | 2,342 | 162,136 | 3,260 |
| I0060 | 1,281 | 84,150 | 18,250 |
| I0061 | 669 | 28,285 | 5,020 |
| I0062 | 765 | 55,002 | 48,105 |
| I0063 | 1,595 | 90,178 | 11,055 |
| I0064 | 2,463 | 142,040 | 65,085 |
| I0065 | 4,079 | 148,019 | 3,000 |
| I0066 | 504 | 44,074 | 5,670 |
| I0067 | 223 | 10,058 | 8,145 |
| I0068 | 3,015 | 129,260 | 97,050 |
| I0069 | 2,461 | 143,314 | 4,130 |
| I0070 | 607 | 41,084 | 30,760 |
| I0071 | 3,840 | 167,286 | 29,935 |
| I0072 | 1,104 | 66,010 | 58,975 |
| I0073 | 1,657 | 76,007 | 11,340 |
| I0074 | 1,997 | 114,017 | 136,985 |
| I0075 | 2,880 | 129,017 | 80,270 |
| I0076 | 1,274 | 54,035 | 5,470 |
| I0077 | 5,790 | 198,264 | 14,880 |
| I0078 | 681 | 57,296 | 40,310 |
| I0079 | 2,661 | 172,052 | 3,795 |
| I0080 | 2,161 | 148,007 | 100,365 |
| I0081 | 1,036 | 50,054 | 167,015 |
| I0082 | 305 | 23,044 | 56,385 |
| I0083 | 2,356 | 105,058 | 16,740 |
| I0084 | 1,651 | 142,106 | 44,190 |
| I0085 | 5,727 | 199,024 | 96,990 |
| I0086 | 2,047 | 94,267 | 9,440 |
| I0087 | 2,852 | 165,050 | 43,145 |
| I0088 | 4,875 | 170,009 | 65,155 |
| I0089 | 1,716 | 81,261 | 21,235 |
| I0090 | 497 | 17,082 | 6,415 |
| I0091 | 619 | 27,024 | 45,395 |
| I0092 | 1,162 | 55,005 | 21,730 |
| I0093 | 331 | 20,123 | 12,145 |
| I0094 | 1,360 | 59,012 | 8,300 |
| I0095 | 3,130 | 113,124 | 42,480 |
| I0096 | 2,207 | 191,021 | 43,270 |
| I0097 | 209 | 11,019 | 16,860 |
| I0098 | 1,248 | 56,023 | 28,580 |
| I0099 | 3,114 | 112,019 | 15,945 |
In the capacity table below, Constraint types is the number of different families of capacity constraints declared by this ficitious airport, e.g. "max 10 departures in 60-minute periods rolling every 15 minutes". Runway constraints and Terminal constraints denote how many of those families of constraints correspond to runway and terminal constraints, respectively. Duration range (min) shows the shortest and longest time window of any capacity constraint in the instance. Rolling period range (min) shows the shortest and longest rolling period of any capacity constraint in the instance. Lastly, Total constraints per day show the number of individual capacity constraints that would result for these synthetic "capacity declarations" in any single day. This number multiplied by 210 days gives the total number of capacity constraints in this synthetic 2020 summer season.
| Instance | Constraint types | Runway constraints | Terminal constraints | Duration range (min) | Rolling period range (min) | Total constraints per day |
|---|---|---|---|---|---|---|
| I0000 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0001 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0002 | 24 | 6 | 18 | 15 - 120 | 15 - 60 | 1,224 |
| I0003 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0004 | 24 | 6 | 18 | 15 - 120 | 15 - 60 | 2,088 |
| I0005 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0006 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0007 | 17 | 5 | 12 | 15 - 120 | 5 - 60 | 936 |
| I0008 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0009 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0010 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0011 | 23 | 5 | 18 | 15 - 120 | 5 - 60 | 1,080 |
| I0012 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0013 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0014 | 17 | 5 | 12 | 15 - 120 | 5 - 60 | 936 |
| I0015 | 17 | 5 | 12 | 15 - 120 | 5 - 60 | 936 |
| I0016 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0017 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0018 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0019 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 648 |
| I0020 | 11 | 5 | 6 | 15 - 120 | 5 - 60 | 792 |
| I0021 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 648 |
| I0022 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0023 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0024 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0025 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 648 |
| I0026 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0027 | 23 | 5 | 18 | 15 - 120 | 5 - 60 | 1,080 |
| I0028 | 24 | 6 | 18 | 15 - 120 | 15 - 60 | 1,224 |
| I0029 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0030 | 17 | 5 | 12 | 15 - 120 | 5 - 60 | 936 |
| I0031 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0032 | 11 | 5 | 6 | 15 - 120 | 5 - 60 | 792 |
| I0033 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0034 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0035 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0036 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 936 |
| I0037 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0038 | 24 | 6 | 18 | 15 - 120 | 15 - 60 | 1,224 |
| I0039 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 936 |
| I0040 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 648 |
| I0041 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0042 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0043 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0044 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0045 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0046 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0047 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0048 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0049 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0050 | 11 | 5 | 6 | 15 - 120 | 5 - 60 | 792 |
| I0051 | 11 | 5 | 6 | 15 - 120 | 5 - 60 | 792 |
| I0052 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0053 | 17 | 5 | 12 | 15 - 120 | 5 - 60 | 936 |
| I0054 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 936 |
| I0055 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0056 | 11 | 5 | 6 | 15 - 120 | 5 - 60 | 792 |
| I0057 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0058 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0059 | 24 | 6 | 18 | 15 - 120 | 15 - 60 | 1,224 |
| I0060 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0061 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0062 | 11 | 5 | 6 | 15 - 120 | 5 - 60 | 792 |
| I0063 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0064 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0065 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0066 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0067 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0068 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0069 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0070 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0071 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0072 | 11 | 5 | 6 | 15 - 120 | 5 - 60 | 792 |
| I0073 | 11 | 5 | 6 | 15 - 120 | 5 - 60 | 792 |
| I0074 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0075 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0076 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0077 | 23 | 5 | 18 | 15 - 120 | 5 - 60 | 1,080 |
| I0078 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0079 | 23 | 5 | 18 | 15 - 120 | 5 - 60 | 1,080 |
| I0080 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0081 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 936 |
| I0082 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 936 |
| I0083 | 17 | 5 | 12 | 15 - 120 | 5 - 60 | 936 |
| I0084 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0085 | 24 | 6 | 18 | 15 - 120 | 15 - 60 | 1,224 |
| I0086 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 936 |
| I0087 | 24 | 6 | 18 | 15 - 120 | 15 - 60 | 1,224 |
| I0088 | 24 | 6 | 18 | 15 - 120 | 15 - 60 | 2,088 |
| I0089 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
| I0090 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0091 | 11 | 5 | 6 | 15 - 120 | 5 - 60 | 792 |
| I0092 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 648 |
| I0093 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0094 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 648 |
| I0095 | 11 | 11 | 0 | 5 - 60 | 5 - 60 | 1,080 |
| I0096 | 23 | 5 | 18 | 15 - 120 | 5 - 60 | 1,080 |
| I0097 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 648 |
| I0098 | 12 | 6 | 6 | 15 - 120 | 15 - 60 | 648 |
| I0099 | 18 | 6 | 12 | 15 - 120 | 15 - 60 | 1,512 |
- Extend this data generation procedure to generate instances considering interactions in a network of airports
- Extend to level 3 airports with the addition of historic status, historic times and historic seat numbers.