Europa Baseline Network¶

This page reviews the configuration of the Europa network and runs a baseline simulation, with all three carriers using the same revenue management system (E). See the Europa network overview for a description of the carriers and their hubs.

import altair as alt
import passengersim as pax
from passengersim.config.checks import check_config

pax.versions()

passengersim 0.94
passengersim.core 0.94

cfg = pax.Config.from_yaml(pax.demo_network("europa"))

Route Maps¶

cfg.plotly_route_map("Brat", center_lon=31, center_lat=52, zoom=2.2)

cfg.plotly_route_map("Stroop", center_lon=31, center_lat=52, zoom=2.2)

cfg.plotly_route_map("Tapas", center_lon=0, center_lat=50, zoom=4.5)

For this example page, we will pre-build the flight connections, so we can do some pre-simulation checking to review our network and verify it is constructed correctly.

from passengersim.connection_builder import prebuild_connections

cfg = prebuild_connections(cfg, inplace=False)

Hub Flight Schedules¶

cfg.fig_hub_schedule("AMS")

cfg.fig_hub_schedule("FRA")

cfg.fig_hub_schedule("MAD")

Check Configuration¶

check_config(cfg)

 ✅ fare classes match carrier booking classes
 ✅ no fares without demands
 ✅ no markets without fares
 ✅ market lowest fares match configured fare prices
 ✅ no demands without fares
 ✅ no demands without paths
 ✅ fare restrictions: 4 unique restrictions found
   - R2: 60 occurrences
   - R1: 176 occurrences
   - R4: 20 occurrences
   - R3: 40 occurrences
 ✅ reference price scaling is consistent
 ✅ choice model restrictions: 4 unique restrictions found
   - r1: {'business': 0.28, 'leisure': 0.3}
   - r2: {'business': 0.4, 'leisure': 0.3}
   - r3: {'business': 0.3, 'leisure': 0.38}
   - r4: {'business': 0.22, 'leisure': 0.22}
 ✅ choice model and fare restrictions match

List of Markets¶

cfg.dataframes.markets

	orig	dest	demand_multiplier	delta_t	distance	lowest_fare
0	KEF	MAD	1.0	None	1794.496	220.0
1	KEF	AMS	1.0	None	1265.292	220.0
2	MAD	AMS	1.0	None	905.703	100.0
3	CDG	AMS	1.0	None	247.307	85.0
4	JFK	AMS	1.0	None	3631.022	800.0
5	KEF	FRA	1.0	None	1490.664	280.0
6	MAD	FRA	1.0	None	880.763	100.0
7	CDG	FRA	1.0	None	277.519	90.0
8	AMS	FRA	1.0	None	227.108	75.0
9	JFK	FRA	1.0	None	3842.424	800.0
10	KEF	HEL	1.0	None	1517.821	380.0
11	MAD	HEL	1.0	None	1829.061	250.0
12	CDG	HEL	1.0	None	1177.052	300.0
13	AMS	HEL	1.0	None	944.658	200.0
14	FRA	HEL	1.0	None	956.086	200.0
15	JFK	HEL	1.0	None	4102.818	1100.0
16	KEF	ATH	1.0	None	2608.413	380.0
17	MAD	ATH	1.0	None	1476.248	200.0
18	CDG	ATH	1.0	None	1309.374	250.0
19	AMS	ATH	1.0	None	1354.762	270.0
20	FRA	ATH	1.0	None	1128.011	220.0
21	JFK	ATH	1.0	None	4925.660	900.0
22	KEF	WAW	1.0	None	1741.344	380.0
23	MAD	WAW	1.0	None	1409.401	260.0
24	CDG	WAW	1.0	None	833.625	210.0
25	AMS	WAW	1.0	None	683.775	150.0
26	FRA	WAW	1.0	None	557.831	170.0
27	JFK	WAW	1.0	None	4252.119	950.0
28	KEF	SIN	1.0	None	7174.082	1400.0
29	MAD	SIN	1.0	None	7067.757	1300.0
30	CDG	SIN	1.0	None	6659.469	1400.0
31	AMS	SIN	1.0	None	6528.644	1100.0
32	FRA	SIN	1.0	None	6384.186	1100.0

Market Pricing Distribution¶

alt.Chart(cfg.dataframes.markets.eval("price_per_mile = lowest_fare / distance")).mark_point().encode(
    x=alt.X("distance", title="Market Distance (miles)"),
    y=alt.Y("price_per_mile", title="Lowest Fare Price per Mile", scale=alt.Scale(zero=False)),
    tooltip=["orig", "dest", "distance", "lowest_fare", "price_per_mile"],
    color="orig",
    shape="dest",
)

Here we have plotted the lowest price fare per mile against the distance. We can see the cluster of lower priced markets with MAD origins (reflecting the LCC that is dominant there) as well as cluster of higher priced markets that are effectively controlled by legacy carriers flying into fortress hubs with zero or nearly zero competition. Note none of this is an outcome of the simulation, this is merely the input/setup of pricing for the network.

Carrier Service Capacity¶

We can calculate the number of seats offered by each carrier on each served route.

cfg.dataframes.legs.groupby(["carrier", "orig", "dest"])[["capacity"]].sum().reset_index()

	carrier	orig	dest	capacity
0	Brat	AMS	FRA	300
1	Brat	CDG	FRA	450
2	Brat	FRA	ATH	450
3	Brat	FRA	HEL	300
4	Brat	FRA	SIN	300
5	Brat	FRA	WAW	300
6	Brat	JFK	FRA	300
7	Brat	MAD	FRA	300
8	Stroop	AMS	ATH	300
9	Stroop	AMS	FRA	200
10	Stroop	AMS	HEL	270
11	Stroop	AMS	SIN	300
12	Stroop	AMS	WAW	170
13	Stroop	CDG	AMS	250
14	Stroop	JFK	AMS	300
15	Stroop	KEF	AMS	320
16	Stroop	MAD	AMS	220
17	Tapas	KEF	MAD	360
18	Tapas	MAD	AMS	240
19	Tapas	MAD	ATH	240
20	Tapas	MAD	FRA	360
21	Tapas	MAD	HEL	120
22	Tapas	MAD	WAW	360

Simulation Results¶

sim = pax.MultiSimulation(cfg)

out = sim.run()

╭─────────────────── Summary revenue statistics · Europa ───────────────────╮
│                            Brat            Stroop             Tapas       │
│ Run      Samples     Mean  ~95% CI     Mean  ~95% CI     Mean  ~95% CI    │
│ Europa       900   932.1K   ±7,237   928.2K   ±7,336   291.7K   ±2,360    │
│ Post-burn-in only · Approx. 95% CI for mean                               │
╰───────────────────────────────────────────────────────────────────────────╯

out.fig_carrier_revenues()

out.fig_carrier_load_factors()

out.fig_carrier_rasm()

out.fig_carrier_yields()

out.fig_fare_class_mix()

out.fig_leg_load_v_distance(carrier="Brat", beeswarm=1000)

out.fig_leg_load_v_distance(carrier="Stroop", beeswarm=1000)

out.fig_leg_load_v_distance(carrier="Tapas", beeswarm=1000)

The Tapas network is much smaller than the other carriers, both in terms of seats flown and also distance of legs, as there are no long-haul legs for this carrier.

Market Shares¶

out.market_shares()

	carrier	Brat	Stroop	Tapas
orig	dest
AMS	ATH	30.679067	69.320933	0.000000
	FRA	55.099152	44.900848	0.000000
	HEL	16.008772	83.991228	0.000000
	SIN	25.466540	74.533460	0.000000
	WAW	16.972072	83.027928	0.000000
CDG	AMS	0.000000	100.000000	0.000000
	ATH	68.816622	31.183378	0.000000
	FRA	92.156420	7.843580	0.000000
	HEL	56.462642	43.537358	0.000000
	SIN	64.801863	35.198137	0.000000
	WAW	79.223803	20.776197	0.000000
FRA	ATH	100.000000	0.000000	0.000000
	HEL	100.000000	0.000000	0.000000
	SIN	100.000000	0.000000	0.000000
	WAW	100.000000	0.000000	0.000000
JFK	AMS	0.000000	100.000000	0.000000
	ATH	58.323946	41.676054	0.000000
	FRA	76.168032	23.831968	0.000000
	HEL	51.752344	48.247656	0.000000
	WAW	52.644019	47.355981	0.000000
KEF	AMS	0.000000	81.250695	18.749305
	ATH	0.000000	42.266403	57.733597
	FRA	0.000000	45.738023	54.261977
	HEL	0.000000	68.014879	31.985121
	MAD	0.000000	0.000000	100.000000
	SIN	0.000000	100.000000	0.000000
	WAW	0.000000	31.917893	68.082107
MAD	AMS	0.000000	33.662542	66.337458
	ATH	12.440618	7.672209	79.887173
	FRA	37.618604	2.587385	59.794011
	HEL	16.600961	22.606935	60.792103
	SIN	55.062873	44.937127	0.000000
	WAW	9.480951	7.467462	83.051586

out.market_shares(agg="orig")

carrier	Brat	Stroop	Tapas
orig
AMS	34.858498	65.141502	0.000000
CDG	65.454048	34.545952	0.000000
FRA	100.000000	0.000000	0.000000
JFK	49.339247	50.660753	0.000000
KEF	0.000000	46.496964	53.503036
MAD	20.158162	14.775521	65.066317

out.market_shares(agg="dest")

carrier	Brat	Stroop	Tapas
dest
AMS	0.000000	63.073160	36.926840
ATH	44.210479	32.516361	23.273159
FRA	48.933506	19.446180	31.620314
HEL	43.679342	37.261557	19.059101
MAD	0.000000	0.000000	100.000000
SIN	50.857332	49.142668	0.000000
WAW	35.785805	20.952654	43.261541