Airline Passenger
Transportation System:
Structure and Dynamics
Lance Sherry
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
Copyright Lance Sherry 2010
1
Airline Passenger Transportation
System (APTS)
• System that provides transportation of
passengers:
– Passengers that provide value at their destinations
(e.g. consult, exchange technology, sales, …)
– Passengers that spend money at their destinations
(e.g. vacation)
• This mode of transportation is characterized by:
–
–
–
–
Service across long geographic distances
Rapid
Affordable
High levels of safety
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
2
Airline Passenger Transportation
System (APTS)
• Transportation accomplished through the interaction of multiple agents:
– Airlines
•
•
•
•
Ticketing
Check-in
Flights
Baggage
– Airports
•
•
•
•
Ground transportation
Security
Terminal services
Airside service (e.g. de-icing, snow removal, …)
– Air Navigation Service Providers
• Navigation infrastructure
• Surveillance function
• Flow Management, Separation, …
– Other supply chain enterprises (rental cars, fuel providers, concessionaires, …
equipment manufacturers)
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
3
Airline Passenger Transportation
System (APTS)
Airline Passenger
Transportation System
1. Quality (i.e. passenger safety)
2. Cost (i.e. airfares + other costs, externalities)
3. Time (i.e. scheduled trip time, reliability = actual trip time)
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
4
(1) Quality of APTS
• mortality risk of passenger air travel
• a passenger chooses a (nonstop) flight
completely at random, what is the probability
that the passenger will be killed during the
flight?
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
5
(2) Cost of APTS
• Airfares + Other Costs of APTS determine
demand for air travel
1. Cost-based pricing
• Direct and Indirect Operating Costs
2. Demand-based pricing
• Supply vs Demand
• Competition
3. Service-based pricing
• Differentiation
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
6
(3) Time & Reliability of APTS
• T = Ground access/egress + Flight time + Schedule
displacement
– Ground Access/Egress
• Airport processing
– Flight Time is Scheduled “block time”
• Includes “schedule padding”
– Schedule Displacement
• Time between when passenger wants to depart/arrive and flight is
scheduled to depart/arrive
• Reliability = Passenger Trip Delays
– Actual arrival time – Scheduled arrival time
– Includes: delayed flights, diverted flights, rebooking for
cancelled flights, missed connections, over-booking
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
7
APTS System Structure
% Direct/Connecting
Itineraries
% Itineraries
Served
Time to Next
Flight
# Airports in HubSpoke Network
Itinerary
Structure
% Passengers
on Direct Itins
# Flights
# Direct Itineraries
# Connecting Itineraries
Seats per Flight
Load Factors
Passengers
Allocated to
Itineraries, and
Itineraries Assigned
to and to Flights
Available Seats for
Rebooking
Total Passengers
Total Pax on Direct Itins
Total Pax on Connecting itins
Direct Pax/Direct itin
Connecting Pax/Connecting iItin
Candidate Itineraries for Rebooking
P(Delayed Flight),
Average Delay for
Delayed Flight
Flight
Operations
P(Cancelled
Flight)
Itinerary
Disruption
P(Direct Itin Delayed)
P(Direct Itin Cancelled0
P(Connecting Itin Delayed)
P(Connecting Itin Cancelled)
P(Connecting Itin Missed_Connection)
Passenger
Trip Delays
Total Pax Trip Delay
Total Direct Itin Pax Trip delay
Total Connecting Itin Pax Trip Dela
% pax On-Time
Average Disrupted Pax Trip Delay
Weather,
Equipment
Failure, …
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
8
Copyright Lance Sherry 2010
Network
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
9
APTS Network
• Network is the manner in which airports are
connected by flights
– Network is a space-time network
– Network determines the itineraries
• Two distinct types of networks
– Point-to-point
– Hub-and-Spoke
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
10
Direct Network
• Each Origin is connected
to each Destination (e.g.
1-2, 1-3, 1-4, 1-5)
• Flights depart Origin in
time to reach Destination
at desired time (e.g. start
of business day)
• # Flights required to
provide service = (n-1)n
– 4* 5 = 20
• # Aircraft = #
Simultaneous Flights = 20
Desired
Arrival
Time
1
2
3
4
5
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
11
Hub-and-Spoke Network
• Each Origin is connected
to each Destination via a
Hub (i.e. 3)
Start of
Day
1
– (e.g. 1-3-2, 1-3-4, 1-3-5)
• Flights depart Origin in
time to reach Destination
at desired time (e.g. start
of business day)
• # Flights required to
provide service = (n-1)2
– 4* 2 = 8
• # Aircraft = #
Simultaneous Flights = 4
2
3
4
5
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
12
Hub-and-Spoke
• Bank
– Flights arrive from Origins at Hub, then depart
from Hub to Destinations
• Typical hub network has 5 – 8 banks per day
– Depends on geography/routes served
• USAirways at CLT vs JetBlue at JFK
• Advantage of Hub – economies of scale
• Disadvantage of Hub – Flight time
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
13
Itineraries
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
14
Itineraries
• Itinerary is the sequence of flights taken by a
given passenger from Origin to Destination
– Direct Itinerary
– Connecting Itinerary
• By definition a given flight (in a hub-and-spoke
network) will have passengers on board
different itineraries
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
15
Direct Itinerary
• Direct Itinerary:
–
–
–
–
–
–
–
Origin
Destination
Scheduled Departure Time Origin (as ticketed)
Scheduled Arrival Time Destination (as ticketed)
Flight Number
D
Flight Seat Capacity
Direct
Type: Direct
Time
H
O
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
16
Connecting Itinerary
Time
D
• Connecting Itinerary
–
–
–
–
Origin
Hub
Destination
Origin-Hub
•
•
•
•
H
Connecting
O
Scheduled Departure Time – origin (as ticketed)
Scheduled Arrival Time - Hub (as ticketed)
Flight Number
Flight Seat Capacity
– Hub-Destination
•
•
•
•
Scheduled Departure Time – origin (as ticketed)
Scheduled Arrival Time - Hub (as ticketed)
Flight Number
Flight Seat Capacity
– Type: Direct
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
17
Actual Itineraries (Hub ATL)
FL_1_NU
T100 Load
ORIGIN DEST HUB AIRLINE DB1B PAX
M
Seats
Factor
FL_2_NUM Seats
DCA
NUL
ATL L
DL
63
1137
143
LGA
ATL DCA
DL
18
1961
134
LGA
ATL DCA
DL
18
1963
134
DCA BHM ATL
DL
27
1137
143
DCA
DL
16
1137
143
1137
0.85314685
143
3
DCA
CHS ATL
TPA ATL
DL
19
0.85314685
3
0.50746268
6
0.50746268
6
0.85314685
3
0.85314685
3
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
T100 Load
Factor
NULL
NULL
1137
143
1137
143
574
146
1164
142
NULL
0.85314685
3
0.85314685
3
0.73972602
7
0.86619718
3
201
0.81592039
8
836
Pax
Flight
Itin Pax
19
10
10
8
5
6
18
Example APTS Network
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
19
Example APTS (5 Airport)
• Markets
1
– Five
– Located in same Time Zone
– Equal distance apart
2
• Transportation Service at
each Market
– Each market has own
airport
– Travel time = 1 Unit Time
between airports
– (e.g. Travel Time 1 to 4 = 4)
3
4
5
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
20
Transportation Demand
• Transportation demand
– Total of 500 passengers
– 100 passengers to each
Destination Market
– 100 passenger from each Origin
Market
– 25 passenger trips from each
Origin market to each
Destination Market
– Passengers are required to be at
Destination for start of day, so
depart with enough time
– Demand for travel at each Origin
to arrive at Destination at start of
day (shown on right)
– 100 pax leave each market
– 100 pax arrive at each
market
Start of Day
at
Destination
Demand to each Market
25
1
5
4
3
2
Time of Day
5
2
4
31
15
24
1
2
35
2
3
4
3
4
5
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
1
Passengers departing 5
to go to 1
21
Direct Flight Network
•
•
•
•
•
Total Passengers = 500
Total Itineraries = 20
# Flights = 4 * 5 = 20
Aircraft Size 25 seats
Distance Traveled =
–
–
–
–
–
–
•
•
•
•
•
4+3+2+1 = 10
3+2+1+1 = 7
2+1+1+2 = 6
3+2+1+1 = 7
4+3+2+1 = 10
40
Total Trip Time = 40
Total Arrival Displacement Time = 0 (all pax
arrive at required time)
Average Trip Time = 40/20 = 2
Max Simultaneous Arrivals at each airport = 4
(at each airport)
Max Simultaneous use of airspace = 5 (at
each TRACON)
Desired
Arrival
Time
1
2
3
4
5
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
22
Direct Flight Network
Origin
Originating
Pax
Destination
Itinerary =
Flights
Pax per
Itineray =
Flight
Trip Time
Arrival
Displacement
1
100
2
3
4
5
1-2
1-3
1-4
1-5
25
25
25
25
1
2
3
4
-
2
100
1
3
4
5
2-1
2-3
2-4
2-5
25
25
25
25
1
1
2
3
-
3
100
1
2
4
5
3-1
3-2
3-4
3-5
25
25
25
25
2
1
1
2
-
4
100
1
2
3
5
4-1
4-2
4-3
4-5
25
25
25
25
3
2
1
1
-
5
100
1
2
3
4
5-1
5-2
5-3
5-4
25
25
25
25
4
3
2
1
-
TOTAL
500
500
60
0
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
23
Hub-n-Spoke Network
•
•
•
•
•
•
Total Passengers = 500
Total Itineraries = 20
# Flights = 4 + 4 = 8
Aircraft Size = 100 seats
Distance Traveled = 12
Total Trip Time =
–
–
–
–
–
–
•
•
1
2
3
Total Arrival Displacement Time =
–
•
•
Flight + Turn + Flight
(5+3+5+6) +
(5+2+4+5)+
(3+2+2+3)+
(5+4+2+5)+
(6+5+3+5)=80
Start of
Day
(some pax arrive earlier than needed)
Average Trip Time = 80/20= 4
Max Simultaneous Arrivals at each airport = 4
(at hub only)
Max Simultaneous use of airspace = 4 (at hub
TRACON only)
4
5
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
24
Hub-n-Spoke: Itinerary Table
Origin
Originating
Pax
Destinatio
n
Itinerary
Pax per
Itinerary
1
100
2
3
4
5
1-3-2
1-3
1-3-4
1-3-5
25
25
25
25
4
2
4
5
1
2
1
-
2
100
1
3
4
5
2-3-1
2-3
2-3-4
2-3-5
25
25
25
25
5
1
3
4
2
1
-
3
100
1
2
4
5
3-1
3-2
3-4
3-5
25
25
25
25
2
1
1
2
1
1
-
4
100
1
2
3
5
4-3-1
4-3-2
4-3
4-3-5
25
25
25
25
4
3
1
4
1
2
-
5
100
1
2
3
4
5-3-1
5-3-2
5-3
5-3-4
25
25
25
25
5
4
2
4
1
2
1
TOTAL
500
500
57
16
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
Total
Trip
Time
Arrival
Displacement
(Early)
25
Hub-n-Spoke: Flight Table
Origin
Destination
Pax per Flight
Total Trip Time
1
3
100
2
2
3
100
1
4
3
100
1
5
3
100
2
3
1
100
2
3
2
100
1
3
4
100
1
3
5
100
2
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
26
Load Factor
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
27
Load Factor
• Load Factor on a flight
– Enplaned Passengers / Available Seats
• Average Load Factor for a set of flights
– *∑ (load factors for set of flights) +/number of
flights
• Network Load Factor
– Total enplaned passengers / Total available seats
– Alternate: RPM/ASM
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
28
Load Factors
• Airlines control Load Factors by:
1. Adjusting Aircraft Size
2. Revenue Management (also known as Yield
Management)
•
•
Computerized Reservation System (CRS)
Internet provides price transparency
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
29
Itinerary Performance
Reliability
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
30
Itinerary Performance
• Passenger Trip Delay = Actual Passenger Arrival
Time – Scheduled (i.e. Ticketed) Arrival Time
• Disruptions resulting in Passenger Trip Delays
1.
2.
3.
4.
5.
Delayed flights
Cancelled flights
Diverted flights
Denied Boarding
Missed Connections
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
31
Direct Itin Disruption
Delayed Flight
• Pax Trip Delay
Probability Pax Trip Delay
Average Delay >
15 mins
– Ticketed Arrival Time –
Actual Arrival Time – 15
min
– DDelayedFlight
• Probability of Pax Trip
Delay
– Probability Flight Delay >
15 minutes
– PDelayedFlight()
15 min
Pax Delay
D
O
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
32
Direct Itin Disruption
Cancelled Flight
• Pax Trip Delay
– Ticketed Arrival Time –
Actual Arrival Time
– DCancelledFlight = f (Frequency
of Service O-D)
• Probability of Pax Trip
Delay
– Probability Flight Cancelled
– PCancelledFlight()
15 min
Pax Delay
D
X
O
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
33
Connected Itin Disruption
Probability Pax Trip Delay
Average Delay >
15 mins
Delayed Flight
• Pax Trip Delay
– Ticketed Arrival Time –
Actual Arrival Time – 15
min
– DDelayedFlight H-D = f
(Frequency of Service O-D)
• Probability of Pax Trip
Delay
– Probability H-D Flight Delay
> 15 minutes
– PDelayedFlight H-D
Pax Delay
15 min
D
H
O
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
Min Connecting Window
34
Connected Itin Disruption
Cancelled Flight (O-H)
• Pax Trip Delay
– Ticketed Arrival Time –
Actual Arrival Time – 15
min
– DCancelledFlight withConnection = f
(Frequency of Service O-H,
H-D)
Pax Delay
D
H
X
O
• Probability of Pax Trip
Delay
Min Connecting Window
– Probability O-H Cancelled
– PCancelledFlightO-H ()
35
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
Connected Itin Disruption
Cancelled Flight (H-D)
• Pax Trip Delay
– Ticketed Arrival Time –
Actual Arrival Time – 15
min
– DCancelledFlight = f (Frequency
of Service H-D)
Pax Delay
D
H
X
O
• Probability of Pax Trip
Delay
Min Connecting Window
– Probability H-D Cancelled
– PcancelledFlightO-H ()
36
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
Connected Itin Disruption
Missed Connection Flight
• Pax Trip Delay
– Ticketed Arrival Time – Actual
Arrival Time – 15 min
– DMissedConnectionFlight = f
(Frequency of Service O-D)
Pax Delay
D
H
• Probability of Pax Trip Delay
– Probability O-H Flight Delay >
15 minutes AND Probability
Pax Misses Connection
– PDelayedFlightO-H *
PMissedConnection()
O
Min Connecting Window
Probability Missed Connection =
Probability O-H is Delayed beyond Min
Connecting Window to H-D
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
37
Probability & Magnitude of Disruption
Itinerary Type
Direct
Type of Itinerary
Disruption
Delayed
Cancelled
Connecting
Delayed
Cancelled
Missed Connection
Probability of Itinerary
Disruption
Based on Probability of
Delayed Flight (typical 0.3)
0.004(Probability of Delay Flight *6.67).
(Typical 0.02)
Based on Probability of
Delayed Flight (typical 0.3)
2 * 0.004(Probability of Delay Flight
*6.67).
Twice probability of
Cancelled Flight (typical 2 *
0.02)
0.1 * Probability of Delayed
Flight.
A
function
of
connecting times and airline
policies regarding holding
flights (typical 0.03)
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
Magnitude of Disruption
(Average)
(Probability
of Delay Flight *6).
10*e
(Typical 60 mins)
Based on Availability of Seats
on subsequent flights and
Time to next flight (average =
300 mins)
10*e (Probability of Delay Flight *6).
(Typical 60 mins)
(5.8902*Load
(0.0483*e
Factor))*Time to Next Flight.
Based on Availability of Seats
on subsequent flights and
Time to next flight (average =
645 mins)
(5.8902*Load
(0.0483*e
Factor))*Time to Next Flight.
Based on Availability of Seats
on subsequent flights and
Time to next flight (average =
38
645 mins)
Rebooking Passengers
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
39
Calculating Passenger Trip Delay
Scheduled
Departure
Time
Schduled
Arrival Time
Seats
# Pax
Flight Status
Delay
Pax Trip
Delay
O-D
06:00
08:00
100
100
Delayed
20 mins
100* 20
O-D
06:10
08:20
120
100
Cancelled
(20 * 130)
+(50 * (210
+40)+
+(30 * 460)
O-H1-D
06:30
10:30
120
100
On-Time
0
O-D
09:30
11:50
150
100
Delayed
O-H2-D
13:00
15:00
120
70
On-Time
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
40 mins
100*40
0
40
Passengers on Cancelled
Flights
(1.8%, Avg 11 hours)
Passengers on Diverted
Flights (0.2%, Avg 3.5
hours)
Passengers Denied Boarding
on Over-sold Flights
(<0.001%)
Passengers on Delayed
Flights (23.9%, Avg 57
minutes)
Passengers On-Time - < 15
Minutes Delay (74%)
The Passenger Trip “Game Wheel”
Not drawn to scale
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
41
Algorithm
Passenger Itin
Load Factor
Flight Performance
For Each Passenger
Itinerary
Direct or
Connecting
Connecting Itin
Direct Itin
• Denied Boarding Not
Shown, same as cancelled
O to H
Cancelled
O to D
Cancelled
O to H
Diverted
O to D
Diverted
Compute Flight Delay
for Diverted Flight
Compute Flight Delay
for Diverted Flight
O to H
Delayed
H to D
Delayed
Missed
Connection
Compute
Pax Delay
for Diverted
Pax
H to D
Cancelled
Compute
Pax Delay
for Diverted
Pax
Rebook Pax O
to D and
Compute Pax
Delays
Rebook Pax H
to D and
Compute Pax
Delay
Compute
Pax Delay
For Delayed
Flight
H to D
Diverted
H to D
Delayed
Compute
Pax Delay
For Delayed
Flight
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
42
Data Sources
• T-100 Domestic Segments Data (U.S. Carriers) –
domestic segment data aggregated by month
• DB1B Coupons Data – a 10% sample of domestic
itinerary data aggregated by quarter
• Flight On-Time Performance Data (ASQP) – daily
on-time arrival data for domestic flights operated
by major U.S. carriers
• FAA Aircraft Registry, which includes seating
capacities by carrier and aircraft type
• Other secondary sources
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
43
Airline Passenger Transportation
System
% Direct/Connecting
Itineraries
% Itineraries
Served
Time to Next
Flight
# Airports in HubSpoke Network
(1) Generate
Itinerary
Structure
% Passengers
on Direct Itins
# Flights
# Direct Itineraries
# Connecting Itineraries
Seats per Flight
Load Factors
(2) Allocate
Passengers
to Itineraries
and to Flights
Available Seats for
Rebooking
Total Passengers
Total Pax on Direct Itins
Total Pax on Connecting itins
Direct Pax/Direct itin
Connecting Pax/Connecting iItin
Candidate Itineraries for Rebooking
P(Delayed Flight),
Average Delay for
Delayed Flight
P(Cancelled Flight)
(3) Generate
Itinerary
Disruption
Performance
P(Direct Itin Delayed)
P(Direct Itin Cancelled0
P(Connecting Itin Delayed)
P(Connecting Itin Cancelled)
P(Connecting Itin Missed_Connection)
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
(4) Compute
Passenger
Trip Delays
Total Pax Trip Delay
Total Direct Itin Pax Trip delay
Total Connecting Itin Pax Trip De
% pax On-Time
Average Disrupted Pax Trip Dela
44
Network Performance Characteristics
•
Transportation System has:
– 16 itineraries
– 500 trips
•
Transportation Service is provided by:
– Network Structure
– Direct Flights vs. Hub-n-Spoke
•
Each Flight has:
–
–
–
–
•
Seat Capacity = SC
Seat Utilization = Load Factor = LF
Likelihood of experiencing delay = P(D)
Likelihood of cancellation = P(C)
Each Trip has Average Trip Delay
– Trip Delay due to Delayed Flight =
DDelayedFlight
– Trip Delay due to Cancelled Flight =
DCancelledFlight
– Trip Delay due to Missed Connection =
DMissedConnection
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
45
Network Performance
• Total Passenger Trip Delays =
Total Passenger Trip Delay from Delayed Flights +
Total Passenger Trip Delay from Cancelled Flights
• Total Passenger Trip Delay from Delayed Flights =
∑
LFOi-Dj*SCOi-Dj * P(D)Oi-Dj* DDelayedFlight Oi-Dj
• Total Passenger Trip Delay from Cancelled Flights
=∑
LFOi-Dj*SCOi-Dj * P(C) Oi-Dj* DCancelledFlight Oi-Dj
i=1,n, j=1,n
i=1,n, j=1,n
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
46
Network Performance
• Under assumption of homogeneous fleet, flight
leg performance ….
–
–
–
–
LFO1-D1 = LFO1-D2 =LFO1-D3 = … = LF
SCO1-D1 = SCO1-D2 = SCO1-D3 = … = SC
P(D)O1-D1= P(D)O1-D2= P(D)O1-D3 = … = P(D)
DDelayedFlight O1-D1 = DDelayedFlight O1-D2 = …. = DDelayedFlight
• Total Passenger Trip Delay from Delayed Flights =
∑
LFOi-Dj*SCOi-Dj * P(D)Oi-Dj* DDelayedFlight Oi-Dj
= #Flights * LF * SC * P(D) * DDelayedFlight
i=1,n, j=1,n
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
47
Performance Metrics
1. % Disrupted Passengers
– Total Disrupted Passengers
• Passengers on Delayed Flights
• Passengers on Cancelled Flights
2. Total Passenger Trip Delay
3. Average Passenger Trip Delay
4. Average Disrupted Passenger Trip Delays
– Average Passenger Trip Delays due to Delayed Flights
– Average Passenger Trip Delays due to Cancelled Flights
– Average Passenger Trip Delays due to Missed
Connections
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
48
Performance: Direct Network
• Total Disrupted Passengers = [P(D) + P(C)] *(#Flights*LF*SC)
• % Passengers Disrupted = P(D)+P(C)
• Total Passenger Trip Delay =
#Flights*LF*SC [(P(D)* DDelayedFlight)+ (P(C)* DCancelledFlight)]
• Average Trip Delay = Total Passenger Trip Delay/#Pax
• Average Disrupted Passenger Trip Delays = Total Passenger
Trip Delay/Total Disrupted Passengers
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
49
Performance: Hub-n-Spoke Network
• Total Disrupted Passengers =
( P(D)H-D * #FlightsH-D * LF*SC) +
(P(D)O-H* P(MC) * #FlightsH-D * LF*SC) +
(P(C)O-H*#FlightsO-H*LF*SC) +
(P(C) H-D*#FlightsH-D*LF*SC)
•
% Passengers Disrupted = P(D) + [P(D)*P(MC)] + 2P(C)
• Total Passenger Trip Delay =
( P(D)H-D * #FlightsH-D * LF * SC *DDelayedFlight ) +
(P(D)O-H* P(MC) * #FlightsH-D * LF*SC *DMissedConnection ) +
(P(C)O-H* #FlightsO-H* LF* SC * DCancelledFlight) +
(P(C) H-D* #FlightsH-D* LF * SC * DCancelledFlight)
•
Average Trip Delay = Total Passenger Trip Delay/#Pax
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
50
Network Performance
Total Pax Trip Delay
( P(D)H-D * #FlightsH-D * LF*SC *DDelayedFlight ) +
(P(D)O-H* P(MC) * #FlightsH-D * LF*SC *DMissedConnection ) +
(P(C)O-H*#FlightsO-H*LF*SC*DCancelledFlight) +
(P(C) H-D*#FlightsH-D*LF*SC*DCancelledFlight)
#Flights*LF*SC [(P(D)* DDelayedFlight)+ (P(C)* DCancelledFlight)]
Direct Itin
30% Connecting Itin/
70% Direct
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
Connecting Itin
51
% Disrupted Passengers
500,000
0.350
400,000
% Disrupted Pax
Total pax Trip Delay (mins)
Total Pax Trip Delay
300,000
200,000
100,000
0
0.300
0.250
0.200
0.150
0.100
0.050
0.000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.1
0.2
% Passengers on Direct itineraries
Total Direct Delay
Total Connecting Pax
0.3
0.4
0.5
0.6
0.7
0.8
% Pax on Direct Flights
Total Pax Delay
% Disrupted Dir Pax
% Disrupted Conn Pax
%Total Disrupted Pax
Average Delay (mins)
Average Pax Delay (mins)
300
250
200
150
100
50
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
% Passengers on Direct Itineraries
Average Delay Direct Pax
Average Delay Connecting Pax
Average Delay (Direct & Connecting) Pax
% Passengers on Direct Itineraries
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
52
0.9
% Disrupted Passengers
500,000
450,000
400,000
350,000
300,000
250,000
200,000
150,000
100,000
50,000
0
0.350
% Disrupted Pax
0.300
0.250
0.200
0.150
0.100
0.050
0.000
0.1
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.2
0.3
0.9
0.4
0.5
0.6
0.7
0.8
% Pax on Direct Flights
% Passengers on Direct itineraries
% Disrupted Dir Pax
Total Direct Delay
Total Connecting Pax
Total Pax Delay
% Disrupted Conn Pax
%Total Disrupted Pax
Average Disrupted Pax Delay (mins)
Average Delay (mins)
Total Pax Trip Delay (mins)
Total Pax Trip Delay
300
250
200
150
100
50
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
% Passengers on Direct Itineraries
Average Delay Direct Pax
Average Delay Connecting Pax
Average Delay (Direct & Connecting) Pax
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
53
0.9
Average Disrupted Pax Delay (mins)
Total Pax Trip Delay
Average Delay (mins)
700,000
600,000
500,000
400,000
300,000
200,000
100,000
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
500
450
400
350
300
250
200
150
100
50
0
0.1
0.9
0.2
0.3
Total Connecting Pax
0.5
0.6
0.7
0.8
0.9
Load Factor
Load Factor
Total Direct Delay
0.4
Average Delay Direct Pax
Total Pax Delay
Average Delay Connecting Pax
Average Delay (Direct & Connecting) Pax
% Disrupted Passengers
% Disrupted Pax
0.300
0.250
0.200
0.150
0.100
0.050
0.000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Load Factor
Load Factor
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
% Disrupted Dir Pax
%Total Disrupted Pax
% Disrupted Conn Pax
54
0.9
Average Disrupted Pax Delay (mins)
Total Pax Trip Delay
600,000
Average Delay (mins)
Total Pax Trip Delay (mins)
700,000
500,000
400,000
300,000
200,000
100,000
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
500
450
400
350
300
250
200
150
100
50
0
0.1
Total Connecting Pax
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Load Factor
Load Factor
Total Direct Delay
0.2
Total Pax Delay
Average Delay Direct Pax
Average Delay Connecting Pax
Average Delay (Direct & Connecting) Pax
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
55
Total Pax Trip Delay
% Disrupted Passengers
0.300
800,000
0.250
% Disrupted Pax
1,000,000
600,000
400,000
200,000
0
0.200
0.150
0.100
0.050
0.000
60
90
120
150
180
210
240
270
300
60
Time to Next Bank (mins)
Total Direct Delay
Total Connecting Pax
90
120
150
180
210
240
270
Time to Next Bank (mins)
Total Pax Delay
% Disrupted Dir Pax
% Disrupted Conn Pax
%Total Disrupted Pax
Average Disrupted Pax Delay (mins)
Average Delay (mins)
700
600
500
400
300
200
100
0
60
90
120
150
180
210
240
270
300
Time to Next Bank
Average Delay Direct Pax
Average Delay Connecting Pax
Average Delay (Direct & Connecting) Pax
Time to Next Bank
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
56
300
Total Pax Trip Delay
% Disrupted Passengers
0.500
1,000,000
0.400
% Disrupted Pax
1,200,000
800,000
600,000
400,000
200,000
0
0.300
0.200
0.100
0.000
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.05
Probability of Delayed Flight
Total Direct Delay
Total Connecting Pax
0.1
0.15
0.2
0.25
0.3
0.35
0.4
Probability of Delayed Flight
Total Pax Delay
% Disrupted Dir Pax
% Disrupted Conn Pax
%Total Disrupted Pax
Average Delay (mins)
Average Disrupted Pax Delay (mins)
400
350
300
250
200
150
100
50
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
Probability of Delayed Flight
Average Delay Direct Pax
Average Delay Connecting Pax
Average Delay (Direct & Connecting) Pax
% On-time
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
57
0.45
Total Pax Trip Delay
% Disrupted Passengers
% Disrupted Pax
1,000,000
800,000
600,000
400,000
200,000
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.500
0.450
0.400
0.350
0.300
0.250
0.200
0.150
0.100
0.050
0.000
0.05
0.45
0.1
Total Direct Delay
Total Connecting Pax
0.15
0.2
0.25
0.3
0.35
0.4
Probability of Delayed Flight
Probability of Delayed Flight
% Disrupted Dir Pax
Total Pax Delay
% Disrupted Conn Pax
%Total Disrupted Pax
Average Disrupted Pax Delay (mins)
Average Delay (mins)
Total Pax Trip Delay (mins)
1,200,000
400
350
300
250
200
150
100
50
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
Probability of Delayed Flight
Average Delay Direct Pax
Average Delay Connecting Pax
Average Delay (Direct & Connecting) Pax
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
58
0.45
Sensitivity Analysis - Total Passenger Trip Delay
Sensitivity Analysis - % Passengers Disrupted
5% Degradation in On-Time
Performance
5% Degradation in On-Time
Performance
One Hour Increase in Time to Next
Flight
One Hour Increase in Time to Next
Flight
10% Increase in Load Factor
10% Increase in Load Factor
10% Shift from Direct to Connecting
Itineraries
10% Shift from Direct to Connecting
Itineraries
-150,000.0000
-100,000.0000-50,000.0000
0.0000
Direct Itin Delay
50,000.0000100,000.0000150,000.0000200,000.0000
-0.060
-0.050
-0.040
Connecting Itin Delay
-0.030
Pax on Direct Itins
-0.020
-0.010
0.000
0.010
Pax on Connecting Itins
Sensitivity Analysis - Average Trip Delay for Disrupted
Passengers
5% Degradation in On-Time
Performance
One Hour Increase in Time to Next
Flight
10% Increase in Load Factor
10% Shift from Direct to Connecting
Itineraries
-100
-50
0
50
100
150
200
Average Trip Delay for Disrupted Passengers
Pax on Direct Itins
Pax on Connecting Itins
Sensitivity Analsysi
CENTER FOR AIR TRANSPORTATION SYSTEMS RESEARCH
59
0.020