An
An Airline’s
Airline’s Block
Block Time
Time Manipulation
Manipulation Within
Within A
A
Congested
Congested Air
Air Traffic
Traffic Environment
Environment
NEXTOR
NEXTOR
National
NationalAirspace
AirspaceSystem
SystemPerformance
PerformanceWorkshop
Workshop
March
14
–
17,
2006
March 14 – 17, 2006
Topics
Topics of
of Discussion
Discussion
• Building Scheduled Block Time For Reliability
• The Impact of Increased Congestion
• Modernization Efforts Which Help To Reduce Delays
• Reducing Scheduling Variability and Cost Through
Improved Planning Processes
• Summary
Building
Building Block
Block For
For Reliability
Reliability
Development
Development Of
Of An
An Accurate
Accurate Scheduled
Scheduled
Block
Block Time
Time Is
Is Critical
Critical For
For Maintaining
Maintaining An
An OnOntime
time Airline
Airline
• Block time is the time from gate departure (brake release) to gate
arrival (brake set). It is composed of:
- Taxi-out time
- Flight time
- Taxi-in time
• Scheduled block times are calculated to achieve a target block ontime :00 for a season. They are based on the historical
performance of a segment when data is available.
Page 1
The
The Building
Building “Block”
“Block” of
of United’s
United’s Operation
Operation
Begins the long term planning process …
How many Pilots do we need?
How many Flight Attendants do we need?
Estimated Block Time Level
(Used in estimating 1 - 2 years out)
How will the banks flow?
Do we have enough aircraft to maintain the schedule?
… And Is Managed Throughout The Aircraft
Scheduling Process
5 months
or more
4 months
3 months
1 to 2 months
Typical schedule development timeline
OA creates
scheduled block
times
Intermediate scheduling
assigns fleets
Current scheduling
works in division
criteria
Further Block Revision Occurs
Based On Bank Structure and
Position of Flights Within Banks
Flight and Onboard
crews scheduled
Schedule
Effective
The
The Impact
Impact of
of Increased
Increased Congestion
Congestion
Stronger
Stronger Passenger
Passenger Demand
Demand And
And The
The Industries
Industries Tendency
Tendency To
To
Answer
Answer These
These Demands
Demands With
With Smaller
Smaller Aircraft
Aircraft Has
Has Resulted
Resulted In
In
A
A Large
Large Increase
Increase In
In Departures
Departures Over
Over The
The Past
Past Few
Few Years
Years
Airline Industry Average Daily Departures By Year
8%
Increase
18,075
17,894
17,858
17,533
17,520
1995 To 1997 Averaged
16,605 Per Day
16,940
16,654
16,717
16,493
16,446
1995
16,390
1996
1997
1998
Note: Departures are from DOT 32 airports
1999
2000
2001
2002
2003
2004
2005
The
The Increase
Increase In
In Departures
Departures Has
Has Contributed
Contributed To
To United
United
Experiencing
Experiencing A
A Higher
Higher Level
Level Of
Of Taxi-Out
Taxi-Out Times
Times
United Airline’s Average Taxi-Out Time Per Flight
13%
increase
18.9
18.5
17.6
16.9
17.4
16.7
16.6
2001
2002
15.6
14.8
1995-97
Average
1998
1999
2000
2003
2004
2005
Comparing
Comparing Similar
Similar City
City Pairs
Pairs With
With Similar
Similar Fleet
Fleet Types,
Types,
United
United Has
Has Experienced
Experienced A
A :02
:02 Minute
Minute In
In Increase
Increase In
In
Flight
Flight Times.
Times. In
In Denver-Chicago,
Denver-Chicago, Flight
Flight Times
Times Have
Have
Increased
Increased :05
:05 Minutes
Minutes
Average DEN-ORD B777 Flight Time
1:52
1:47
1995-1997
2004 - 2005
~5
minute
increase
While
While The
The Percent
Percent Of
Of United’s
United’s Overall
Overall Flow
Flow Control
Control
Delays
Delays Have
Have Remained
Remained Constant
Constant …
…
United Airline’s Flow Control Delays At The Gate (Percent of Flights)
3.5%
2.5%
2.8%
2.5%
2.8%
2.6%
3.1%
2.4%
1.7%
1995
1996
1997
1998
1999
2000
2001
2002
2.4%
2.1%
2003
2004
2005
.. .. .. The
The Intensity
Intensity Of
Of The
The Average
Average Flow
Flow Control
Control Gate
Gate
Delay
Delay Has
Has Increased
Increased
United Airline’s Average Length Per Delay And Flow
Control Delays At The Gate (Percent of Flights)
49.4
1995 To 1997 Averaged 34
Minutes Per EDCT
36.8
46.4
43.5
42.8
42.0
~12
minute
increase
30.2
3.5%
2.5%
45.7
42.7
35.2
2.8%
51.2
2.5%
2.6%
1998
1999
2.8%
3.1%
2.4%
2.6%
2002
2003
2.4%
1.7%
1995
1996
1997
2000
2001
2004
2005
Air
Air Traffic
Traffic Delays
Delays Have
Have Created
Created Significant
Significant Cost
Cost for
for United
United
Examples of Congestion Cost –
Mid-1990s to present
Impact on United
3.7 Minutes of increase in taxi out time
per Flight
$62.8 Million
2 minute increase in route time per flight
$52.6 Million
12 minute increase in flow control gate delay
(based on 2.5% of Flights Per Day)
$4.2 Million
$119 million
of increased
costs
NOTE: Costs Based on ATA Numbers; Gate Delay = $24, Taxi-out Delay = $31 and Enroute Delay = $48
Modernization
Modernization Efforts
Efforts Which
Which Helps
Helps To
To
Reduce
Reduce Delays
Delays
Although
Although Technology
Technology Enhancements
Enhancements Help,
Help, Runways
Runways
Remain
Remain the
the Most
Most Effective
Effective Capacity
Capacity Enhancement
Enhancement
Estimated Percent Planned Capacity Improvement (VFR Day)
Based On The FAA’s Capacity Benchmark Study
Planned Runway Capacity
5
31
Atlanta
5
35
Houston
4
29
Minneapolis
40
O’Hare*
Los Angeles
Technology Enabled Capacity
6
11
none
none
San Francisco none
27% average
increase
6% average
increase
Short
Short Term
Term Technical
Technical and
and Procedural
Procedural Changes
Changes Will
Will Also
Also Help
Help
• SOIR – Simultaneous Operations on Intersecting Runway
Operations at ORD:
– Increases runway arrival capacity
• SOIA – Simultaneous Offset Instrument Approaches at SFO*:
– Allows dual runway operations in IFR conditions
• Idle descent - Working with DEN and ORD centers and
TRACONs:
– Optimal descent at flight idle saves fuel and environment
* 10-15% of operating days will benefit from SOIA/PRM
Reducing
Reducing Scheduling
Scheduling Variability
Variability And
And Cost
Cost
Through
Through Improved
Improved Planning
Planning Processes
Processes
Increasing
Increasing The
The Number
Number Of
Of Block
Block Seasons
Seasons
Over
Over The
The Past
Past Two
Two Years,
Years, B:00
B:00 Trends
Trends By
By Month
Month Have
Have
Shown
Shown Similar
Similar Patterns
Patterns During
During The
The Winter
Winter 22 And
And Summer
Summer 11
Seasons
Seasons
Winter 2 (Jan-Apr): Between Jan and Apr, B:00
tends to increase as winter head winds diminish
75.0%
B:00 OT Percent
70.0%
2006
65.0%
2005
60.0%
2004
Summer 1 (May-Aug): Typically
experience a drop in May and then a
gradual increase over the summer
55.0%
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
50.0%
Current Seasonal Breakdown:
Winter 2
Summer 1
Fall
Winter 1
To
To Understand
Understand The
The Impact
Impact Of
Of Block
Block Seasons
Seasons On
On Overall
Overall
Performance,
Performance, We
We Trended
Trended Out
Out The
The Following
Following City
City Pairs
Pairs
B737 – Omni Directional
ABQ-DEN
DEN-DFW
DEN-MCI
DEN-MSP
DEN-OMA
DFW-ORD (N/S)
MSP-ORD (N/S)
ORD-CMH
ORD-LGA
SEA-SFO (N/S)
SFO-LAX (N/S)
SLC-DEN
(N/S) = North/South City Pair
A320 – Omni Directional
LAX-MCO
LAX-ORD
ORD-DCA
ORD-LGA
SAN-ORD
SEA-ORD
SFO-LAX (N/S)
SFO-ORD
SFO-SAN (N/S)
B757 – Omni Directional
DEN-LGA
DEN-ORD
LAX-ORD
ORD-BWI
PDX-ORD
SEA-ORD
SFO-EWR
SFO-LAX (N/S)
SNA-ORD
Based
Based on
on These
These City
City Pairs,
Pairs, Each
Each Month
Month Tends
Tends To
To Vary
Vary From
From
The
The Annual
Annual Average
Average In
In Terms
Terms Of
Of Flying
Flying Time
Time And
And Taxi-out
Taxi-out
Time
Time
Flight Time (Monthly Variance) - East Bound
B737 - East
10.0
B757 - East
Taxi-out (Monthly Variance) - West Bound
A320 - East
B737 - West
10.0
8.0
B757 - West
A320 - West
8.0
6.0
Minute Variance
Convective Weather
4.0
2.0
0.0
-2.0
-4.0
6.0
Convective Weather
4.0
2.0
0.0
-2.0
-4.0
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
-10.0
Feb
-8.0
-10.0
Feb
-6.0
-8.0
Jan
-6.0
Jan
Minute Variance
Dec
Jan
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
Taxi-out (Monthly Variance) - East Bound
Nov
-8.0
-10.0
Oct
-6.0
Sep
-4.0
Aug
0.0
-2.0
Jul
2.0
A320 - West
(Head Wind Impact)
10.0
8.0
6.0
4.0
2.0
0.0
-2.0
-4.0
-6.0
-8.0
-10.0
Jun
Minute Variance
4.0
B757 - West
May
(Tail Wind Impact)
6.0
Minute Variance
B737 - West
Apr
8.0
Flight Time (Monthly Variance) - West Bound
A320 - East
Mar
10.0
B757 - East
Feb
B737 - East
Block Minute Variance From Annual Average
By
By Combining
Combining Both
Both Directions
Directions Along
Along With
With Actual
Actual Flight
Flight Time
Time
Variance
Variance and
and Actual
Actual Taxi-out
Taxi-out Variance,
Variance, The
The Trend
Trend Below
Below
Illustrates
Illustrates The
The Monthly
Monthly Overall
Overall Variance
Variance From
From The
The Annual
Annual
Average
(Minute Variance
Variance From
From Annual
Annual Average)
Average)
Average (Minute
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
Larger Variation Has
Resulted In Shorter Block
Seasons
Smaller Variation Has Resulted In
Longer Block Seasons
-0.6
-0.8
-1.0
Jan
Feb
Mar
Winter 2
Apr
May
Jun
Jul
Summer 1
Aug
Sep
Oct
Fall
Nov
Dec
Winter 1
By
By Adjusting
Adjusting The
The Block
Block Seasons,
Seasons, It
It Is
Is Possible
Possible To
To Further
Further
Minimize
(Minute Variance
Variance From
From
Minimize The
The Month
Month To
To Month
Month Variation
Variation (Minute
Block
Block Season
Season Average)
Average)
Block Minute Variance From Seasonal Avg
0.5
Wind/Taxi Optimized Season
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
Current Season
-0.3
-0.4
Jan
Wind/Taxi Optimized
Season:
Current Season:
Feb
Mar
Winter 2
Winter 2
Apr
Spring
May
Jun
Summer 1
Jul
Aug
Summer 2
Summer 1
Sep
Oct
Nov
Dec
Fall
Winter 1
Fall
Winter 1
While
While The
The Wind/Taxi
Wind/Taxi Optimized
Optimized Seasons
Seasons Will
Will Impact
Impact The
The
Current
Current Aircraft
Aircraft Scheduling
Scheduling Periods,
Periods, Adding
Adding A
A Spring
Spring Season
Season
Will
Will Reduce
Reduce Monthly
Monthly Variability
Variability And
And Maintain
Maintain The
The Aircraft
Aircraft
Scheduling
(Minute Variance
Variance From
From Block
Block Season
Season Average)
Average)
Scheduling Periods
Periods (Minute
Block Minute Variance From Seasonal Avg
0.5
Schedule Optimized Season
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
Current Season
-0.3
-0.4
Jan
Schedule Optimized
Season:
Current Season:
Feb
Mar
Winter 2
Winter 2
Apr
May
Spring
Jun
Jul
Aug
Summer 1
Summer 1
Sep
Oct
Nov
Dec
Fall
Winter 1
Fall
Winter 1
Next
Next Steps:
Steps:
• Determine the implications of adding an April only and
breaking June/July within the Aircraft Scheduling process
• Understand the cost of an additional schedule period
• Decide to use either the wind/taxi block seasons or the
schedule period block seasons
• Implement additional seasons
Scheduled
Scheduled Block
Block Time
Time Optimization
Optimization
Based
Based On
On The
The Summer
Summer Season
Season (May-August),
(May-August), Currently
Currently 80%
80%
Of
Of Flights
Flights Are
Are Within
Within +/-:05
+/-:05 Minutes
Minutes When
When Comparing
Comparing
Scheduled
Scheduled And
And Actual
Actual Block
Block Times
Times
140
On 357 Flights, We Could Have
Taken At Least 1 Minute Of
Scheduled Block Time And Still
Achieve A Time At The 65th% …
120
Number of Flights With
Exact Scheduled Block
and 65th% Time Block
100
Number of Flights
… However, We Needed At Least
1 More Minute To Achieve The
65% Time On 509 Flights
80
60
Possible Area For Block
Reduction
40
20
0
-16 -14
-12 -10
-8
-6
-4
-2
0
2
4
6
8
10
12
Scheduled Minute Variance From Actual 65th%
14
16
18
20
22
24
Variability Between Flights Allows For Actual Minute
Variations From Scheduled While Still Maintaining A B:00
Of 65%
For A Given B:00 Performance, Difference
Between Scheduled And Actual 65% Can
Range Up To 3 or 4 Minutes
100
95
90
85
B:00 % On-Time
80
75
70
65
60
55
50
Variation In Range When Comparing
Scheduled Block and Actual 65% Block At
B:00 65%
45
40
35
30
25
20
-16 -15 -14 -13 -12 -11 -10 -9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
7
8
9
10
Scheduled Minute Variance From Actual 65th%
Actual Was Less Than Scheduled
Actual Was More Than Scheduled
11
12
13
14
15
16
Depending
Depending On
On Specific
Specific City
City Pair
Pair Congestion,
Congestion, A
A Given
Given
Market
Market Will
Will Have
Have A
A Completely
Completely Different
Different Block
Block
Distribution
Distribution When
When Compared
Compared To
To Other
Other Markets
Markets
Denver To Billings – Actual 2005
Block Time Cumulative Distribution
80
60
80% of Flights Fall Within
A :15 Minute Period
40
20
0
60
70
80
90
100
110
120
O’Hare To La Guardia – Actual 2005
Block Time Cumulative Distribution
Actual Block Minutes
100
Percent of Flights
Percent of Flights
100
80
60
40% of Flights Fall Within
A :15 Minute Period
40
20
0
100
110
120
130
140
Actual Block Minutes
150
160
Scheduled
Scheduled Block
Block Time
Time Optimization
Optimization Became
Became Effective
Effective With
With
The
The January
January Schedule
Schedule Change
Change And
And Has
Has Been
Been Implemented
Implemented
Through
Through The
The Summer
Summer Schedule
Schedule Period
Period
Jan 9 – May 3, 2006
May 4 – Sep 5
Sep 6 – Oct 30
Ted Markets Only
All Markets Included
All Markets Included
• O’Hare Inbound Protected
• O’Hare Inbound Protected
• Recommend ORD Inbound
Protection
• +/-:05 Minute Change Limit
• +/-:03 Minute Change Limit
(short haul – less than 5 hours)
• +/-:05 Minute Change Limit
(short haul – less than 5 hours)
• +/-:05 Minute Change Limit
(long haul – 5 or more hours)
• +/-:10 Minute Change Limit
(long haul – 5 or more hours)
• No Change to International Tags
• No Change to International Tags
th
• 50 Percentile Lower Bound
• 50th Percentile Lower Bound
th Schedule
Since
Since The
The January
January 99th
Schedule Change,
Change,
Optimization
Optimization Has
Has Improved
Improved Actual
Actual B:00,
B:00, But
But Has
Has
Only
Only Made
Made A
A Small
Small Impact
Impact On
On Arrival
Arrival Performance
Performance
With Optimization
Without Optimization
80.0%
Percent of Flights
75.0%
70.0%
67.7%
65.7%
65.0%
60.0%
55.8%
55.0%
54.9%
50.0%
Block OT :00
NOTE: Based on actual performance between Jan 9 – Mar 5, 2006
Arrival :00
Flights
Flights With
With Optimization
Optimization Changes
Changes Of
Of 33 Or
Or More
More Minutes
Minutes Tend
Tend
To
To Have
Have The
The Largest
Largest Impact
Impact
+/- 1 To 2 Minutes
Impact of Block Optimization
3.5%
+/- 3 To 5 Minutes
3.3%
3.0%
2.5%
2.0%
1.5%
1.3%
1.0%
0.5%
0.2%
0.3%
0.0%
Block OT :00
NOTE: Based on actual performance between Jan 9 – Mar 5, 2006
Arrival :00
Improved
Improved Airline
Airline Management
Management For
For Flights
Flights
Impacted
Impacted By
By A
A Ground
Ground Delay
Delay Program
Program
Objective:
Objective:
Improved
Improved EDCT
EDCT Compliance
Compliance By
By Maintaining
Maintaining Aircraft
Aircraft On
On The
The Gate
Gate
----
Crews
Crews Are
Are Paid
Paid When
When Brakes
Brakes Released
Released
Reduced
Reduced Fuel
Fuel By
By Continuing
Continuing To
To Use
Use GPU’s
GPU’s
Helps
Helps To
To Reduce
Reduce Airport
Airport Congestion
Congestion
Analysis:
Analysis:
To
To Identify
Identify The
The Impact
Impact of
of Flow
Flow Control
Control (FC)
(FC) Delays
Delays On
On Taxi-out
Taxi-out
Performance
Performance
-- Included
Included All
All UAX
UAX Flights
Flights Arriving
Arriving Into
Into ORD
ORD Between
Between 9/05
9/05 –– 2/06
2/06
-- Compared
Compared The
The Period
Period Between
Between 9/05-12/05
9/05-12/05 With
With 1/06-2/06
1/06-2/06
While
While Weather
Weather Was
Was Better
Better Than
Than Average
Average Over
Over The
The Past
Past Two
Two
Quarters,
Quarters, United
United Express
Express Flights
Flights Still
Still Experienced
Experienced A
A Ground
Ground
Delay
Delay Program
Program 10%
10% of
of The
The Time
Time
10% of all flights have "FC" delays
GDP impacted flights
Regular Flights
On
On Days
Days With
With GDP’s,
GDP’s, The
The Overall
Overall Taxi-out
Taxi-out Time
Time Increases
Increases By
By
55%
55% For
For Flights
Flights Arriving
Arriving Into
Into O’Hare
O’Hare
Taxi-out increases by 7.8 minutes during GDP days when
compared to regular days
Minutes Per Flight
25
21.9
20
14.1
15
10
5
0
Taxi-out during GDPs
Taxi-out without a GDP
Grouping
Grouping UAX
UAX Stations
Stations Into
Into 44 Categories
Categories Based
Based On
On
The
The Taxi-out
Taxi-out Difference
Difference Between
Between “FC”
“FC” and
and “Non-FC”
“Non-FC”
Flights
Flights (From
(From Sept
Sept 11 –– Dec
Dec 31,
31, 2005)
2005)
Incremental Taxi-out
Less than 5 minutes
Incremental Taxi-out
Between 5 and 7.5 mins.
Incremental Taxi-out
Between 7.5 and 10 mins.
Incremental Taxi-out
Greater than 10 minutes
AVP
BNA
BOI
CAE
CLE
DTW
GSP
JAX
LAN
LNK
MLI
OKC
PIT
RAP
TVC
XNA
YWG
YYC
ABE
ATW
AUS
AZO
BHM
CAK
CID
CLT
FAR
FSD
FWA
MEM
MSP
OMA
RSW
SAT
SAV
SDF
SGF
SYR
ALB
BTV
BUF
CHS
COS
CVG
DAY
DSM
GRB
ICT
LEX
MBS
MCI
MSN
PIA
RIC
SBN
SPI
ABQ
ATL
BDL
BMI
CMH
CRW
CWA
GRR
GSO
HPN
IAH
IND
MDT
MHT
MKE
MSY
MYR
ORF
PVD
PWM, RDU, ROA
ROC, STL, TUL
TYS, YOW, YUL
[ORD inbounds only]
If
If We
We Focus
Focus Only
Only On
On Jan
Jan And
And Feb
Feb 2006
2006 Data,
Data, We
We See
See A
A
Clear
Clear Decrease
Decrease In
In The
The Incremental
Incremental Taxi-out
Taxi-out Minutes
Minutes
Incremental Taxi-out
Less than 5 minutes
Incremental Taxi-out
Between 5 and 7.5 mins.
Incremental Taxi-out
Between 7.5 and 10 mins.
Incremental Taxi-out
Greater than 10 minutes
ABE
ATW
AVP
AZO
BNA
BTV
CAK
CLE
CLT
CRW
CVG
CWA
FSD
HPN
LAN
LEX
MBS
MLI
MSP
OKC
OMA, PIA, PWM,
ROC, SAV, XNA,
YEG, YWG, YYC
AUS
BHM
BUF
CAE
DSM
DTW
GRB
GSP
IND
JAX
LNK
MDT
MEM
MHT
MSN
PVD
RDU
ROA
SDF
SGF
YUL
ATL
BMI
CHS
CID
FAR
FWA
GRR
ICT
MCI
MKE
MSY
PIT
RIC
STL
SYR
TUL
TYS
ABQ
ALB
BDL
CMH
COS
DAY
GSO
IAH
MYR
ORF
SAT
SBN
SPI
TVC
YOW
[ORD inbounds only]
To
To Summarize…
Summarize…
•• The
The Industry
Industry Is
Is Experiencing
Experiencing Increased
Increased Traffic
Traffic And
And Higher
Higher Load
Load
Factors
Factors Which
Which Limits
Limits The
The Options
Options To
To Cancel
Cancel Flights
Flights
•• Modernization
Modernization Continues
Continues To
To Be
Be The
The Key
Key Element
Element In
In Reducing
Reducing
Delays
Delays By
By Increasing
Increasing Airport
Airport Capacity
Capacity
•• In
In Addition,
Addition, Understanding
Understanding The
The Schedule
Schedule Structure
Structure And
And Options
Options
Within
Within An
An Airlines
Airlines Control
Control Offers
Offers Further
Further Opportunities
Opportunities To
To
Minimize
Minimize Variability
Variability