Quick Change Market Case Study Airbus A321 PDF
Quick Change Market Case Study Airbus A321 PDF
Quick Change Market Case Study Airbus A321 PDF
12-1994
This thesis is brought to you for free and open access by Embry-Riddle Aeronautical University – Daytona Beach at ERAU Scholarly Commons. It has
been accepted for inclusion in the Theses - Daytona Beach collection by an authorized administrator of ERAU Scholarly Commons. For more
information, please contact commons@erau.edu.
NOTE TO USERS
132
UMI
AIRBUS A320/321 QUICK CHANGE MARKET ANALYSIS
- A CASE STUDY
by
December 1994
UMI Number: EP31959
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations
and photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
UMI'
UMI Microform EP31959
Copyright 2011 by ProQuest LLC
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106-1346
11
A CASE STUDY -
by
THESIS COMMITTEE:
H:ovs/^4^(o—
Hans Indlekofer
H/?/H
Dead of Faculty, DaytonyBeach campus Date
IV
ACKNOWLEDGMENTS
I would like to thank my thesis chairman Dr. Boris Trnavskis not only for his
continuous qualified support and critical review during the preparation of the thesis, but
also for the skills I was able to obtain by taking his courses at Embry-Riddle. The energy
he put into his courses motivated and provided me with the skills necessary to complete a
Also, I would like to thank the people of Deutsche Aerospace Airbus, Airbus
Industrie and the airline that supported me aside from their daily routines. In this context,
I want to give special attention to Hans Indlekofer from Deutsche Aerospace Airbus who
generously made this thesis possible by allowing me to conduct the research within the
company, and to the airline's network planning manager who spend a considerable
Year: 1994
The purpose of this thesis is to evaluate and compare the Boeing B737 QC to the
Airbus A320/321 QC aircraft, and to determine their relative market within a sample
airline. The technical design of the two Airbus aircraft in a mixed QC operation were
considered with respect to the requirements of a particular airline. Direct operating costs
and payload range data for all three aircraft were calculated.
linear programming fleet planning model was developed that considers more than the
direct operating costs of a particular aircraft. The cost components included were direct
capital costs of the conversion, costs of positioning flights at low load factors, conversion
The model was then applied to an actual network and potential new routes. The
results are presented and analyzed. The outcome is considered the potential market for
TABLE OF CONTENTS
LIST OF ABBREVIATIONS ix
LIST OF TABLES x
LIST OF FIGURES xi
1 INTRODUCTION 1
1.1 Problem Statement 2
2 RESEARCH METHOD 9
2.1 Aircraft Comparison and Evaluation 9
4.2 Loading/Unloading 24
4.4 Scheduling 26
7 INPUT PARAMETERS 70
7.1 The network 70
9 CONCLUSION 94
REFERENCES 96
LIST OF ABBREVIATIONS
USD: US Dollars
x
LIST OF TABLES
Table 1 -A320/321 QC Advantages and Disadvantages of Different Mam Cargo Door Locations 16
Table 6 — B737, A320, A321 Payload Range Data and Differences among the Different Versions and Types40
Table 7 -B737, A320, A321 DOC per SKO and DOC per SKO Differences 44
Table 8 - B737 QC, A320 QC, A321 QC DOC per SKO and DOC per SKO Differences Used in Passenger
Configuration 45
Table 9 - B737 QC, A320 QC, A321 QC DOC per TKO in Cargo Configuration 48
Table 10 — Leg Designators, Distances, Initial Demand and Opportunity Costs of the Network 73
Table 13 — Average Costs per Cost Component, Aircraft Type, and Time Period 79
Table 19 - Summary of Input Parameters for the DOC per SKO and TKO Calculation 99
XI
LIST OF FIGURES
Figure 6 B737-300 QC, A320-200 QC, and A321-100 QC payload range diagram in passenger
configuration 36
Figure 7 B737-300 QC, A320-200 QC, and A321-100 QC payload range diagram in
cargo configuration 37
Figure 11 B737, A320, and A321 DOC per SKO in normal and quick change-passenger configuration 43
Figure 12 B737, A320, and A321 cash costs per SKO in normal and quick change-passenger configuration 46
Figure 13 B737 QC, A320 QC, and A321 QC DOC per TKO in cargo configuration 47
Figure 14 B737 QC, A320 QC, and A231 QC cash costs per TKO in cargo configuration 48
1 INTRODUCTION
The quick change (QC) aircraft concept was originally developed by the Boeing
Company in the early 1970's. The "quick change aircraft" is a rebuilt passenger aircraft.
Within about 45 minutes, it can be converted from an all passenger aircraft with only
belly cargo space, to an all cargo aircraft with no passenger seats available and the
To date, only the B737 and B727 can be rebuilt to QC versions. Airbus Industrie,
however, is developing a QC version of its Airbus A320 and A321 aircraft. Now, airlines
that wish to convert some of their passenger aircraft will have a choice between Airbus
A320/321 QC and to determine the market of the A320/321 QC for a sample airline.
operations. This may include a substitution of wide-body aircraft on night mail routes by
QC aircraft. The present cargo/mail traffic volumes on some wide-body routes exceed the
capacity of the B737 QC aircraft and would require parallel operation of two or more
advantageous for the airline, because the A320/321 QC has a higher capacity than the
B737 QC.
1
A-Air is a hypothetical airline modeled on a major European airline.
2
This thesis examines the technical feasibility and market for the Airbus A320/321
Quick Change aircraft compared to the Boeing B737 QC aircraft in the present and
planned European cargo and night-mail network of A-Air. A time horizon of twelve
The analysis involves assessing the economic feasibility which is defined as the
degree to which converting and operating A320/321 aircraft would produce cost savings
compared to converting and operating B737 aircraft. The economic feasibility considers
the costs of the conversion but does not include an analysis of financing alternatives.
European cargo and night-mail network is defined as all flight itineraries for which
A-Air schedules narrow body cargo aircraft (quick change or cargo aircraft), as well as
positioning flights with passengers on board. This also includes destinations outside
1993, Moss conducted initial research concerning an A320/321 QC.2 In this study, he
investigated broad benchmarks and requirements for a freighter and quick change aircraft.
Without detailed technical solutions, a broad aircraft definition was proposed and
presented.
calculated the optimum load density and maximum payload for both versions. Further, he
determined and compared payload range diagrams for the B737-200 QC and the B757
package freighter, and the center of gravity movements during loading and unloading for a
front and aft main cargo door position. Those calculations, however, were only of limited
value to the current study. First, the calculations are based on weight estimates that were
unrealistically low and on parameters that were not necessarily true for A-Air (e.g.
MTOW). Therefore, the maximum payload might have been too optimistic. Second, the
B737-200 aircraft is not the main competitor of the A320/321. The newer B737-300
2
Hermann Moos, "Zukunftige Einsatzbedingungen und Anforderungen an ein Airbus A320/321
Frachtflugzeug in der Serien- und Umrustlosung," (Future conditions and requirements to an Airbus
A320/321 freighter as a series and conversion aircraft), Diplomarbeit FH Wiirzburg September 1993
3
Walter Borchard, A320 Feasibility Study. Deutsche Aerospace Airbus, February 1994.
4
series should be considered. Third, no direct operating cost comparisons were included in
the study.
In February and March 1994, Kwik and Sprenger prepared another study on the
A320 QC which was mainly concerned with the cabin layout, the cargo loading system,
and the required R&D effort needed for the cabin design.4 Although the basic seat
configuration of A-Air was used as a base, the study noted that the A320/321 QC had to
be redesigned to make it competitive with the B737 QC. First, they modified the design
to include a crash net to comply with safety regulations. This net, however, would cost a
container position and leave the A320 QC with the same container capacity as the B737
QC. Therefore, a 9-g system comparable to the B737 QC system had to be implemented.
Second, the height of the loading system was unacceptably high and led to an aisle width
reduction of 3.6 inches which may cause passenger service problems. Third, only the
A320 but not the A321 was considered in the study. Finally, some details such as seat
Literature pertaining the B737 QC was mainly supplied by A-Air. Since this
thesis compares the two aircraft types, the appropriate parts of the B737 literature will be
presented in the main body of the thesis. Basic economic data associated with the B737
4
Wilfried Sprenger and Karl Kwik, A320 QC Cabin Layout Deutsche Aerospace, March 1994.
QC were taken from a study the airline prepared before acquiring the aircraft. Technical
The Boeing aircraft as used by the airline were converted by Pemco Aeroplex Inc.,
a major supplier of cargo conversion kits. The basic aircraft considered in this analysis is
In 1965, Hiat and Plewes5 studied potential advantages of the B737 QC and B727
QC. They mentioned the advantage of lower capital costs associated with higher aircraft
utilization as one major advantage of the QC concept. The study predicts a high demand
In the past, several mathematical models have been used to solve aircraft fleet
five U.S. national carriers and found as one major conclusion that these airlines do not
necessarily make full use of fleet planning models during the acquisition process. Models
5
MA. Hiatt and K.C. Plewes, The Quick-Change Convertible Cargo-Passenger Aircraft Will Aid
Air Freight Development in the Next Decade, (Seattle: Boeing Co., 1965. Document Number 650782).
6
Robert H. Hammer, "Fleet and Airplane Acquisition Planning of Regional Airlines" (M.S. Thesis,
Massachusetts Institute of Technology, 1983).
6
Furthermore, he is not aviation specific but considers other transportation modes. His
To date, many different fleet planning models have been developed. Simple
models may consider only one period and portray reality in simplified terms. Kirby8 and
Wyatt , for example, assumed a single type fleet with known demand and the constraint
that all demand must be met either with the fleet vehicles or by outside hire. Other early
models use linear programming algorithms to optimize fleet planning10. These early
efforts, however, are of limited use since the lack of computer resources forced them to
the objective function of the model. It allows performing sensitivity analysis if input
Schick and Stroup use a computer supported model developed by the Douglas
expressed either as direct operating costs, capital costs, or a combination of these. The
fleet mix is determined by several computer supported steps with a human analyst
involved in each step. Carriage of passengers and cargo is considered in the model.
the assumption that cost minimization is the only true objective for fleet planning since
price setting is considered beyond the control of a particular airline. This model is
resale value of an aircraft into account and assumes somefixedcosts with introducing a
new aircraft type at an airline. All variables are considered to be time dependent.
viewpoint. He does not detail all the variables affecting fleet planning but looks at a
complete route system served by several airlines. Similar routes are classified into a
11
James C. Goodboy and James G. Gilbertson, Freighter Network Analysis Model. (Seattle: Boeing
Co., 1960)
12
GJ Schick and JW Stroup, "Experience with a Multi-Year Fleet Planning Model", The International
Journal of Management Science 9 (1981): 389-96.
13
DP Shube and JW Stroup, Fleet Planning Model, (Sacramento: Douglas Aircraft Company, 1975),
Paper 6440.
14
Christopher Colin New, "Transport Fleet Planning For Multi-Period Operations", Operational
Research Quarterly 26 (1975): 151-166.
15
Armando C. Silva, Cell Fleet Planning: An Industry Case Study. (Cambridge: Massachusetts
Institute of Technology, Department of Aeronautics and Astronautics, Flight Transportation Laboratory,
May 1984), FTL Report R84-4.
certain number of cells which reduce the complexity of the model. These cells share
common characteristics such as stage length and passenger volume. He uses his fleet
planning model to forecast demand for new aircraft - not only for one airline, but for a
objective function.
Lockheed Co. takes a more macroscopic view by analyzing total cargo systems.17
Certain aspects such as identification of major cost elements were helpful to identify
16
Jeph Abara, "Applying Integer Linear Programming to the Fleet Assignment Problem", Interfaces 19
(July/August 1989): 20-28.
17
R.B. Ormsby, Development of Total Airline Profit Model Program to Permit Simulation and
Evaluation of Total Air Cargo System, (Georgia: Lockheed-Georgia Co., 1969), SAE TRANS 690413.
9
2 RESEARCH METHOD
A two step analysis was employed to determine the market for A320/321 QC
aircraft within A-Air and evaluate the competitiveness of the Airbus aircraft compared to
the B737 QC. In the first step (chapters 3-5) the characteristics of the new aircraft types
were analyzed and compared to the existing B737 QC. This refers to a technical,
In the second step (chapters 6-8), results from the first step and airline data were
used to simulate the impact of the availability of three aircraft types in the network on the
minimum cost fleet mix. This was accomplished by formulating and solving a linear
programming fleet planning model. The two steps are further explained in the two
sections below.
Initially, the technical differences between the three aircraft were outlined. This
was accomplished by comparing technical papers and documents obtained from the
airlines and airframe manufacturers and discussing the technical layout with Airbus and
airline engineers. Since the technical layout of the two Airbus aircraft was still in the pre-
10
planning phase, the layout of the Airbus aircraft was adapted to the specific requirements
of A-Air to the maximum possible extent. An inductive approach was used with the
purpose of identifying and quantifying technical benchmarks of the Airbus aircraft that
managers; and presenting them information about the A320/321 QC in the form of
technical drawings and data. The Airbus layout and its technical design were discussed
with respect to the characteristics of the daily operation within A-Air. This procedure
identified the aspects of a mixed QC operation which might be different from a single
type operation.
The economic comparison was performed independent of the route structure but
employed a standard method of aircraft cost comparison. Three configurations for each of
the three aircraft (B737, A320, A321) were compared: the aircraft as a normal passenger
aircraft in cargo configuration. The method used for economic analysis consists of three
steps. Initially, the operating empty weight (OEW) of the aircraft, the structural weight
limitations, and the aircraft configuration was specified referring to the technical
specification. In the second step, payload range data were calculated. The performance
information of step two was then combined with analytical (e.g. fuel) and empirical (e.g.
handling fees) cost data from an A-Air DOC-calculation software to determine costs per
seat-km (SKO) or ton-km (TKO) for an aircraft that is operated at full payload over a 500
11
To determine the optimum fleet mix and thus the potential market for A320/321
network using network cost minimization as the objective function. The major cost
components, for different mixes of aircraft types, for each leg of the network were
identified and quantified. Development of these cost components over time was then
forecast using assumed growth rates for the input parameters that determine these costs.
Rates and parameters were taken from the results of chapter 5, supplied by the airline, or
estimated.
suggested fleet mix, for each year of the planning horizon. The schedule complies with
aircraft scheduling constraints which are imposed by general aircraft scheduling and QC
specific requirements. Optionally, initial stock of a specific aircraft type, and aircraft
availability.
18
ISA atmosphere, 150 NM alternate, 30 min. holding, 5% contingency fuel.
12
The LP model was then developed and processed using SAS/OR software. The
solution to the LP was critically evaluated and the results were considered the potential
All aircraft are equipped with a cargo door for main deck container loading. The
door of the B737 QC is located in the front section of the fuselage, whereas the A320/321
QC will have the door in the aft section. Figure 1 shows the location of the cargo doors.
The A320/321 QC main cargo door (MCD) will have a dimension of 86"xl42" compared
to 84.6" x 123" for the B737 QC. The larger cargo door gives more flexibility in sizing
the seat pallets. It will be possible to design the seat pallets with four instead of three
rows per pallet. This reduces the problem of loose carpet borders at the pallet edges,
because the number of pallets is reduced by one compared to the B737 QC. The seat
pallets will be further discussed in section 3.3. Also, larger seat pallets reduce the
conversion time because the ground crew has fewer bolts to unscrew.
In the B737 QC there is no choice of cargo door locations because the aircraft is
only available with a front cargo door. Also, an aft door location would not be possible,
because the fuselage is too short and the loading equipment would interfere with the wing
tips. The Airbus aircraft have different fuselage dimensions and are still in the design
phase. Therefore, the cargo door location can still be changed. Advantages and
14
84.6"x123"
B737 QC
A320 QC
disadvantages of the different door locations for the Airbus aircraft are summarized in
Table 1
From a Deutsche Aerospace Airbus (who will be responsible for the QC retrofit)
design standpoint, a front door location is disadvantageous. Each of the main fuselage
sections is designed and completed by the respective manufacturer before they are joined
together in the final assembly line. If design changes (such as a cargo door) on aircraft
sections are needed after an aircraft has been completed (e.g. a QC retrofit), each
company involved in manufacturing the affected section has to be involved in the design
change. The front door would be partially located in the fuselage section that is
of the design such as door design, relocation of affected aircraft systems (wiring, etc.),
production process planning, time planning and cost planning. This additional
coordination effort could be avoided if the door is located in the aft, because in this case it
would be located completely in the Deutsche Aerospace Airbus section. However, if the
door is located in the aft, it will be partially located in the noncylindric section of the
fuselage. This will make the door design more complicated plus the aft door location
makes the relocation of affected aircraft systems more difficult (e.g. hydraulic lines).
Also, the aft fuselage encounters higher aerodynamical and structural forces which makes
the door about 100-200 kg heavier and the design more expensive.
The B737 QC does not have these problems, because the retrofit is performed by a
single company (Pemco) which is licensed by Boeing and has sole responsibility for the
16
conversion. Also, the aircraft is not composed of sections that were designed, completed
and equipped, with all aircraft systems beforehand, by independent companies as it is the
From an operational standpoint, the aft location is preferred. First, the center of
gravity will not be within take-off limits if the door is located in the front. That means on
ferry flights, weight would be required. Second, during loading and unloading only an aft
location will provide sufficient load on the front wheel, which is especially critical for the
longer A321 QC. This assumes a standard loading procedure where each container is
moved to the frontmost position before the next container is loaded into the aircraft
(further discussed in section 4.2). The B737 QC has no threat of tail skipping due to the
different fuselage size. Third, the threat of engine damage during loading and unloading
is reduced with an aft door location, because the loader does not have to move in front of
the engine inlets. Interference with the wing tips is not critical.
A major problem associated with any cargo door is the aversion of passengers to
sit next to it. In case the door is located in the front (as is in case of the B737 QC) mainly
first and business class passengers are sitting next to it. This can be avoided if the door is
moved to the back which would also reduce the air noise caused by the door.
Additionally, an aft door would offer the prospect of offering the A320/321 as a combi
aircraft comparable to the principle of the B747 combi. With a front door, a combi
19
Although technically possible, the required cabin layout for combi operation with afrontdoor location
is not accepted by A-Air.
18
From an economic standpoint, the higher weight of the aft door will increase the
fuel consumption and make the aircraft less fuel efficient as it could be. This effect,
however, will be partially offset because a more aft location of the center of gravity
(weight of the door in the back) is aerodynamically advantageous (lower angle of attack;
less downdraft required by the stabilizer to control stability of the aircraft) and the
So far, from the perspective of A-Air and Airbus, an aft door location is preferable
and its advantages outweigh the disadvantages. Therefore, an aft cargo door location as
shown in Figure 1 represents the current planning status (the A321 QC will have a similar
with occasional fluid leaks causing cabin and passenger soiling led to a retrofit with an
Airbus will offer several options for a cargo loading system (CLS) that will be
comparable to the 9-g20 system currently installed in the B737 QC. Customers will have
a choice between 1V4", l3/4", and 2" system height above the seat rails. Currently, the
B737 QC system has a height of 1%". Therefore, the Airbus aircraft offer the option of a
V2 system height reduction. The seat pallets add an additional 1" height similar to the
B737 QC. Single and double row CLSs will be available. The double row system is
necessary for night mail operation. If desired, power drive units can be installed in the
cabin. However, they will add additional weight to the conversion with marginal benefit.
The B737 QC was initially equipped with electrical systems. They proved to be very
delicate and failed several times causing the electrical drive to block the rolls and making
To avoid the disadvantages of the higher cabin floor, Airbus is presently reviewing
the possibility of integrating the CLS into the seat rails. This would reduce the system
height (including seat pallet height) to \lA" and significantly reduce the slope of the ramp
in the cabin. If this reduced height system can not be installed, there will be no significant
The cabin layouts of the aircraft under study are shown in Figure 2. Due to the
increased floor height, the seats next to the overwing emergency exits have to be
removed. This reduces the seating capacity by four seats in the A320 and two in the
B737. This does not affect the A321 because it has a different design for the emergency
exits. The front lavatory of the A320 has to be moved forward by 18", because otherwise
it would not be possible to load the ninth container. The middle lavatory of the A321 has
,I 71=
^JOIf®-:©!*© 1
-©-:®1:©!?©^
-B737 QC-
A320 Q C
MODIFIED POSITION
MHtosmam
iiraiiiiiiSiE
A321 QC
Figure 3 shows the cross section of the Airbus cabin. The cabin isle width
depends upon the height of the CLS. The isle width of the unconverted aircraft is 21".
With a system height of 3 " isle width is reduced to 17.4" which will impose service
problems during daytime operation, although still within legal limits. In case of the B737
QC, the floor height increase does not cause a reduction in isle width because even with
higher seats there is still enough spacing between the sides of the back rests and the cabin
wall. The vertical clearance for the standard 125" x 88" 9-g container is sufficient.
The seat pallets will have a width of 125" similar to the B737 QC to fit into the
seat vans. To accommodate the full width of the cabin floor, 3" wide rails will serve as
side guidance for the seat pallets and container. It will be surfaced with rubber or plastic
matching the carpet design. The length of the seat pallets can be variable and will be
Integrating the conversion into the aircraft systems affects mainly Aerospatial
components in the cockpit. This aspect was not yet reviewed, but differs considerably
from the integration of the conversion in the case of the B737. The systems software has
to be adapted (different weight, door warning, etc.) to integrate the new configuration into
the electronic centralized aircraft monitoring system (ECAM). This is not necessary in
the case of the B737, which is not does not have a comparable system. Further analysis of
this aspect of the conversion is not practical because it is very technical and involves to be
SMOKE DETECTOR"
Although the A320/321 QC will have the MCD in the aft section of the fuselage,
there will be no major differences in the conversion procedure. A standard ground crew
of four to five people, specified in the ground handling agreement, will perform the
conversion. It takes about 20 minutes for the B737 QC to convert the aircraft after the
last passenger has left the aircraft. Initially, catering removes all trolleys and containers
from the galleys. Simultaneously, two people open the seat pallet locks, unplug the
wiring for the floor path marking system, and stow the movable class divider. As soon as
the 1L stairway can be removed, one loader opens the cargo door and the seat van is
brought into position. Two door seal protection devices are put in place before the
loading/unloading begins. The seat pallets are removed through the main cargo door and
are stowed in two seat vans. The maximum seat pallet width that can be stowed in the
seat vans and that can be handled with the container loader is 125". In the case of the
A320/321 QC, a third seat van will be necessary to stow all the seat pallets. The vans are
heated to keep the seats at a comfortable temperature. After the seat pallets are removed,
the aircraft is ready for loading. In the case of the A320/321 it will be necessary to install
protection walls between the main deck cargo compartment and the front and aft galley.
24
The conversion back to the passenger version is done in reverse order. At the end,
however, the floor path marking has to be checked and signed in the technical logbook.
4.2 Loading/Unloading
The B737 QC is loaded from the front. One container at a time is lifted into the
aircraft and then moved manually by one loader into the rear position where it is secured
by YZ-locks. The next container can not be loaded into the aircraft until the loader has
secured the rear container and returned to the front position of the aircraft. There is not
enough space between the container used by A-Air and the aircraft sidewalls to pass a
container in the cabin. The same will be true for the A320/321 QC. Only one container
at a time can be loaded into the aircraft, but into the frontmost position of the main deck.
Since the Airbus will have an aft MCD location, the risk of engine damage
especially during winter operation is reduced, and the stairway IL does not interfere with
the container loading equipment and can remain at the aircraft. However, the seat vans
will need a second door in the backside of the truck because they can no longer approach
the aircraft parallel to the longitudinal axis but have to approach the fuselage at a 90
degree angle. This problem might be avoided if a container loader is positioned between
the aircraft and the seat van. Figure 4 illustrates the position of the loading equipment
during cargo operation for the A320/321 QC. Please note that the seats are not
25
necessarily stowed on a pallet train and that some airports do not allow pallet train
operation as shown in Figure 4. There, the pallet train is located outside the aircraft area
and a special transporter picks up one container at a time and carries it to the container
Most European airports are currently able to fully handle B737 and A320/321
aircraft. Handling ability here refers to the availability of appropriate loading equipment,
certified ground personal, and whether the airport may be used by the respective aircraft
type. The aircraft under study meet Stage 3standards. Therefore, noise restrictions would
affect them in the same way if night curfews become an operational problem.
The A320/321 require two additional lower deck container loaders compared to
the B737 which does not have lower deck containers and is therefore loaded manually.
Other than that, if the Airbus aircraft were added to A-Air's QC fleet, no significant
4.4 Scheduling
The aircraft schedule has to be balanced.21 This means that the first flight in the
evening after the conversion to a cargo airplane, has to be the same aircraft type as the last
flight before the conversion back to a passenger aircraft in the morning. For example,
from a practical operational standpoint, if the first cargo leg outbound from a conversion
station (after the conversion to cargo configuration) is operated by a B737 QC, then the
last inbound cargo leg (before the conversion back to a passenger aircraft) to the same
station must be a B737 QC. It cannot be served by an A320 QC for example. Also, the
21
Refer to chapters 6.3 and 6.4 for further discussion about aircraft balance.
27
number of aircraft of a specific type departing from a particular station has to be the same
as the number of aircraft arriving at this station. Otherwise the schedule will result in the
If the aircraft type that flies on a certain route varies over time, the number of
available 9-g containers at each station has to be adjusted according to the aircraft
capacity, because the number of container positions is different for each aircraft type.
Therefore, if the B737 QC is replaced by a larger aircraft such as the A321 QC, the
aircraft will have to carry empty containers or the empty containers will have to be carried
by truck.
28
characteristics of the basic aircraft, the B737-300 or A320/321 in this case, the additional
weight of the conversion, the capacity in cargo and passenger configuration, and the costs
of the conversion including additional costs for structural weight increases (MTOW,
MLAW, MZFW). Characteristics such as an improved cargo loading system, an aft cargo
door position, and other technical design features, where the A320/321 may offer
potential advantages, were not valued in the direct operating cost (DOC) calculations
below.
performance, and basis aircraft price. Table 2 shows the structural weights that were used
throughout the analysis. It has to be noted that the weight limits shown for the Airbus
aircraft are not yet available to airlines. Airbus Industrie, however, is reviewing the
29
technical feasibility of these new structural weight limitations. Since preliminary analysis
showed that both aircraft would only be competitive with higher limits, it was assumed
that the weight increase would be included in the conversion. In the case of the A320, the
weight limit increase will probably be achieved by service life reductions, which can not
yet be specified. Additionally, the take-off rating was increased to 26,500 lb. The A321
will require technical design changes. It was assumed that the aerodynamic performance
is not affected by the conversion except for the impact of the higher OEW. The
aerodynamic performance such as speed and fuel consumption were taken from Airbus22
and Boeing23 manuals and will not be presented in further detail at this point. The B737
QC performance data were increased by 1% to account for the difference beween the
Boeing manual and actual A-Air operational experience with the aircraft. The
corresponding adjustment for the Airbus aircraft is 3%. Both of these adjustments reflect
22
Airbus Industrie, Performance Doc. P2210 Rev. 2, June 93 and Performance Doc. P21131 Rev.l,
May 92.
23
Boeing ^ ™ ™ ^ ; * i Aircraft rnmpanv. Performance Doc. D6-37042-4, Nov. 14 1984.
30
the experiences of A-Air with the reliability of performance data supplied by the
manufacturers.
Basic aircraft prices are USD 37 million for the B737, USD 47 million for the
A320 (both with CFM 56 engines) and USD 55 million for the A321 (with IAE engines),
all in A-Air specification. These prices are guidelines only because exact prices are
confidential and negotiable, and will vary depending on how the aircraft is equipped.
Also, the actual price may vary considerably depending upon the number of concessions
granted by the manufacturer to a particular airline. The aircraft price, however, will not
affect the cash costs as presented later in the analysis. The conversion costs are not
The conversion to a quick change aircraft adds additional weight to the OEW.
Two cases have to be considered: the new OEW of the quick change aircraft in passenger
mode and the new OEW of the quick change aircraft in cargo mode. To date, Deutsche
Aerospace Airbus can not provide weight estimates for the quick change conversion.
Therefore, the additional weight was estimated by extrapolating the additional weight of
the B737 PEMCO conversion. The weight of the individual components was subdivided
into variable weight components (weight varies with the aircraft size; e.g. seat pallets) and
fixed weight components (weight does not vary with aircraft size; e.g. cargo door). It was
31
assumed that the weight of the variable components would vary in a linear manner with
the number of container positions. In the case of the A320 QC, there is an additional
fixed weight increase of 200 kg due to MTOW limit increase. The A321 QC will require
determine the maximum structural payload. In the case of the cargo configuration, the
OEW has to be corrected by removable cabin interior and the cabin crew. Additionally,
weight conservatism is included in the calculation. The amount used in the calculations is
standard A-Air conservatism. It counts for weight increases during operation due to
repairs, dirt, etc. Additionally, the Manufacturer's Empty Weight has to be corrected in the
32
case of the A321, because the aircraft was actually lighter than stated by the manufacturer.
The B737 QC offers a capacity of eight 88" x 125" 9-g containers plus additional
bulk space in the belly. The A320 QC has nine, and the A321 QC twelve 88" x 125"
container positions (refer to Figure 2 on page 20). The Airbus aircraft offer seven (A320)
and ten (A321) AKH container positions for the lower deck. The B737 has bulk capacity
only. It was assumed that the Airbus aircraft can not be bulk loaded, because bulk loading
is an option for Airbus aircraft but is not available to A-Air. The main disadvantages of
the containers for cargo operation are their weight and their size. The containers are
included in the OEW and therefore reduce the maximum net payload by 560 kg for the
The AKH container is smaller than the wide-body LD-3 container which is the
respective lower deck container for wide body aircraft. Therefore, A320/321 lower deck
containerized cargo has to be reloaded into LD-3's to optimize space utilization if the
cargo continues in wide body aircraft.24 Additionally, the volume utilization of LD-3
To determine the payload, the structural payload has to be compared to the volume
limited payload as shown in Table 5. The structural payload is defined as the difference
between the MZFW and the OEW.25 The volume limited payload is calculated by
multiplying the available cargo volume with the average cargo density (or by adding the
24
Technically, AKH's can be carried in wide body aircraft. However, the containers do not fit into the
cargo compartment in an optimum manner.
25
In case the difference of the MLAW minus MZFW is less than standard reserves, the structural
payload may be less (landing weight limited). In the case of the A320/321, however, this is not the case
34
Table 5.-- Maximum Payloads for Different Aircraft Types and Configurations
average weight of the container and passengers). A standard weight of 1,700 kg for the
main deck container, 500 kg for the lower deck container, 84 kg per passenger, and 14 kg
baggage per passenger was used. Every 35.7 passengers utilize one lower deck container
Due to the higher cabin floor in passenger configuration, passenger seats beside
the overwing emergency exits have to be removed. The seat structure may not project
into the emergency exit. Therefore, the B737 seating capacity is reduced by two, and the
A320 seating capacity is reduced by four seats. The A321 has a different emergency exit
Deutsche Aerospace has not yet published any prices for a quick change
conversion. Initial internal cost calculations (based on full costs) have also not provided a
solid basis for price estimates. Therefore, it was assumed that the conversion could be
offered at market prices which were estimated by the Deutsche Aerospace sales
different aircraft types. The conversion price for an A320 was fixed at USD 3.5 million
and the one for the A321 at USD 4.0 million. This includes also the costs for structural
weight increases. If the Airbus aircraft are not competitive with the B737, a reduction of
the basic aircraft price to reduce the capital costs might be considered by Airbus Industrie.
Final study prices are USD 40 million for the B737 QC, USD 50.5 million for the A320
The payload range data for all aircraft were calculated using the same method.
Figures 5 to 10 inclusive show payload range diagrams for the studied aircraft types and
configurations. Refer to Table 2 and Table 4 for associated aircraft weights. The
-i r -, r r
-> r -i r -i
"• -, r - -, r -. r -,
A320-20ff -« >- -i
^ ^
r
^-
-i
^P"t- m
r -i
p ^
: -r —
J L J L J c J - 1 1 - J t-
•7i7 j n n J i. J
16 000
**VCJCC/WPar
1 r n r -. r ^ X ^ -, r - * * ^ r -. r
Q
£
-J L J L , . , c J« -it. J
1 r 1 r -i r i r -t -> r -i r •y r -, T ^ K -> r i
-J k. J L J 7!> , .
:
— ? s p ^ Alternate
: |SACqndi^op« r
: S^Cpntlpgerpy-
1 r -i r i T r -> r . r . r ,
: ^ T r -i
-J <- J L J - -, t- : -. u
B<ng»0cm)
-; : • . . ; *
:- • -
20 000
; A321 l t O C C f t v
:r m
\G<&tZrtx*y£*>
» :
; ^ ^*i - *
15 000 - Aj20-20p QC H »
- : r
!l747400-dc>a( "• -» -•
u. _. J i: 1 COt.ClfdLPtx****' a x u
10 000
^, k
>
:- -; T r r
:
150 Nto Alternate
5 000
&A Conditions
5*% Contingency
J -5 * - ': ;
|- a u u
RaipOan)
Figure 6. B737-300 QC, A320-200 QC, and A321-100 QC payload range diagram in
passenger configuration
r r
;A«l'-1<WOCCcroo H -
• •"
m ml J .1 a 4 i, 1
' L C U
«r
T r r
^ r ""
t» " '
: "" - '**^ 1 L : j
>3}0>200 QCTOcuo
* •<*
"NJ^ T r r T r
B?37*MK> OO Oagp —-—
r -. T r ^"**'««^ 1 J - , •
r r ' Nr r
n r
. .^^
;: ::
;-
J
S-
i-
*""- - ^ ^ -
**""^ VW
-i i r ". r r r T r r T T r
- r- -, ,
• •- -- - r- -. 1 r r T r r- 1 r r
:
0
-. -. r T r r -, T r T r r T Tr
Figure 7. B737-300 QC, A320-200 QC, and A321-100 QC payload range diagram in
cargo configuration.
- - . . .
:" " ": *- • . . :
20 000
r r
r r r 1 • ^ T
: n r r
-, .
„I „--...,.,»
%
*w * *
1747400- -i -i - mr
-, r r
3XC*gCfT«*TPf*t " ^ ^ K
* *• *: , i u
1 J
: L. J
* u
10 000
, ^
• r ~> i r r -i •
•"^i 1 r r
!»7*,,,,»«iN^
;- _J -1
. L. -> J
R<ng»0on)
r i r r -i
1
— 1 — 1 — 1 — 1 — 1 _ 1 1 • 1 ** : :.: : ^
;AA20-200_QCCa^o
r
-AJ2O-20Q "* "* , -, T r r j - -
, * •-^,','ls,,,%^^
[ -] J i i_
V*
r I T
:
:
150 Htf Alternate
-ISA Conditions
r r- -» •
! -51S Contingency^ "
:c : ... :
RcngiOcm)
. ^ • • ^ j ^ a ^ i i L L & U -> ]
V^iVsc^ ']']'
:*. . : . • .
u J \ [ !! :
000
r r * •
: • •
• r - i i r r -> •
*»»
000
r T. r r 1 f r T r r -. r r
. i- J
150 NM4lternate
000 ISA Conditions
r 1 r r "I
; 5%Cot«lnflenty:
|-
:, L . :
in the same configuration. It can be seen that the A320 has an advantage on routes above
2,000 km because the B737 is unable to carry the maximum payload. The A321 has
Table 6 presents the results in detail and shows the differences between the B737
configuration, QC aircraft in cargo configuration). Delta values in the columns show the
differences between the Airbus aircraft and the B737 in the same configuration, whereas
delta values in the rows show the differences with respect to the non-converted aircraft.
The former is used to evaluate the additional capacity of a larger aircraft while the latter is
used to evaluate the loss of capacity due to the additional weight of the conversion.
more payload at 1,251 km (74%) higher optimum range. If the aircraft is operated at
maximum range, the difference amounts to 3.41 (49%). Respective values for the A321
are 5.6 t (38%) at 654 km (39%) higher range and 7 t (101%) at maximum range.
maintain their payload advantage relative to the B737 QC (absolute payload advantage
decreases slightly). Values are 2.01 (14%) for the A320 QC and 5.11 (37%) for the A321
QC. Range differences shift slightly. The A320 QC increases its optimum range
advantage at full payload to 1,091 km (57%) whereas the A321 advantage is reduced to
317 km (17%). At maximum range, both Airbus aircraft lose some of their payload
advantage. The A320 QC offers only 2.7 t (39%) more payload and the A321 QC 5.5 t
(17%). The maximum range advantage (disadvantage A321) remains almost unchanged.
40
Table 6.-- B737, A320, A321 Payload Range Data and Differences among the Different
All aircraft are still able to carry maximum passenger load. Remaining cargo capacity
advantage of 2.3 t (14%) at a 1,091 km (57%) higher optimum range. Values for the
A321 QC are 6.11 (39%) and 317 km (17%) respectively. At maximum range, the
26
Difference of maximum payload minus number of passenger seats times 98 kg. May vary slightly in
case the remaining volume limits the remaining capacity.
41
payload advantage of the A320 QC is 2.3 t (23%) and of the A321 QC 6.2 t (64%).
Comparing different versions of the same aircraft type (B737 see Figure 8, A320
see Figure 9 A321 see Figure 10), the A320 QC in passenger configuration looses the
least compared to the unconverted aircraft. The A320 QC looses 11 (6.4%), the B737 QC
11 (7.3%), and the A321 QC 1.6 t (8%). The optimum range increases slightly or
remains constant due to the increases in the MTOW for all aircraft. As a rough guideline,
300 km can be subtracted from the optimum range for each ton decrease in MTOW. At
maximum range, the B737 QC looses 90 kg (1%), the A320 QC 792 kg (5%), and the
A321QC 1.6t(8%).
The required take-off field length for the A321 QC at ISA, sea level, and MTOW
has increased by 5.3% from 2,200 m to 2,316 m, and by 9.6% from 2,865 m to 3,139 m
under ISA +20° conditions and 2,000 ft pressure altitude. Landing field length has
without a thrust increase from present levels since most runway lengths exceed these
values.27
The respective values for the A320 QC change slightly, because the T/O rating
was increased from 25,000 lb. to 26,500 lb. to avoid performance problems. Required
take-off field length at ISA has decreased by 7% from 2,195 m to 2,042 m, and by 9%
from 3,231 m to 2,926 m under ISA + 20° conditions. Landing field length increases
27
A321 performance is currently insufficient in a few specific weather conditions at some airports in the
network. This is the case for 83 t and 85 t MTOW. However, thrust increases for these exceptions are
normally not considered by the airline. According to Airbus Industrie, there are airlines that operate the
A321 with the higher take-off weight without a thrust rating increase
5.2.2 Direct Operating Costs (DOC) per SKO
Using the above calculated payload range data, direct operating costs were
calculated using a standard A-Air computer program (all values in USD per SKO). This
program utilizes the following method. For aircraft comparison, a standard stage length
of 500 NM is used. The payload range data supply the appropriate input parameters such
as block fuel, block time, payload, and available seats and freight for each aircraft type at
this stage length. An annual yearly utilization of 1,920 flights per year (about 8.5 block
hours per day) is assumed. Based on theses figures, annually offered seat-km, ton-km,
block fuel, and block hours were calculated. Additional details are provided in Appendix
A.
Direct operating costs per SKO are separated into variable and fixed cost
components. Fuel costs were calculated using a price of USD 0.218 per liter. This
costing method assumes that the aircraft consumes the whole block fuel on a trip, which is
normally not true. However, for the purpose of aircraft comparison, this method is
acceptable. Maintenance costs are separated into airframe and engine maintenance.
These cost components are a function of the aircraft weight and type and are based on
empirical studies. Landing, handling, and navigation charges are a function of the aircraft
weight and the payload (handling charges) of the aircraft. The fixed cost components are
technical, capital, insurance, and cockpit/cabin crew costs. The capital costs are shown as
This costing method treats all aircraft as if they were flown in the same
configuration during the entire year. It does not yet show the DOC of a quick change
43
aircraft that is flown in a mixed operation. If such a mixed operation increases the annual
utilization (i.e. more flights per year in either cargo or passenger configuration), the fixed
costs will be spread over more flights with a subsequent reduction of the DOC per seat-
km or per ton-km. This effect, however, will influence the Boeing and Airbus aircraft in
the same way. The direct operating costs are shown in Figures 12 through 15.
Figure 11 shows total direct operating costs per SKO for the passenger versions of
Figure 11. B737, A320, and A321 DOC per SKO in normal and quick change-passenger
configuration.
44
the B737, A320, and A321. Respective individual values and cost differences are shown
in Table 7 and Table 8. The A320 is about 11% more fuel efficient than the B737 and has
about 15% lower crew costs per SKO. Respective values for the QC version are 7.5%
and 14%. However, higher landing fees (11%) and capital costs (10.4%/11.8%) eliminate
the DOC advantages of the A320. The A321 is significantly cheaper in fuel, handling,
navigation, fixed technical, and crew costs. The high aircraft price is spread over more
seats and therefore the capital costs are not significantly higher. Total DOC per SKO are
Table 7.--B737, A320, A321 DOC per SKO and DOC per SKO Differences
(US cents per SKO)
B737 A320 Delta B737 Delta B737 A321 Delta B737 Delta B737
(ln%) (ln%)
Fuel Cost 0.826 0.738 -0.088 -10.63% 0.662 -0.165 -19.92%
Airframe 0.573 0.567 -0.007 -1.17% 0.492 -0.081 -14.14%
maintenance
Engine maintenance 0.165 0.166 0.001 0.44% 0.218 0.053 32.30%
Landing fees 0.454 0.504 0.051 11.19% 0.451 -0.003 -0.65%
Handling fees 1.244 1.194 -0.049 -3.97% 1.126 -0.118 -9.48%
Navigation charges 0.701 0.684 -0.018 -2.54% 0.575 -0.127 -18.04%
Total variable costs 3.963 3.853 -0.110 -2.78% 3.523 -0.440 -11.09%
Fixed technical costs 0.850 0.837 -0.013 -1.54% 0.721 -0.129 -15.21%
Capital costs 2.335 2.578 0.243 10.40% 2.386 0.051 2.17%
Aircraft 1.194 1.311 0.116 9.74% 1.214 0.019 1.62%
deDreciation
Spares 0.105 0.123 0.019 17.69% 0.113 0.009 8.41%
DeDreciation
Aircraft interest 0.953 1.046 0.093 9.75% 0.968 0.015 1.61%
Spares interest 0.083 0.098 0.015 17.93% 0.090 0.007 8.57%
Insurance 0.050 0.055 0.005 9.63% 0.051 0.001 1.55%
Cockpit crew 0.637 0.544 -0.093 -14.58% 0.430 -0.206 -32.42%
Cabin crew 0.483 0.413 -0.070 -14.58% 0.421 -0.062 -12.87%
Total fixed costs 4.356 4.427 0.071 1.63% 4.009 -0.347 -7.96%
Total direct costs per 8.319 8.280 -0.039 -0.47% 7.533 -0.786 -9.45%
SKO
45
Table 8 - B737 QC, A320 QC, A321 QC DOC per SKO and DOC per SKO Differences
Used in Passenger Configuration.
(US cents per SKO)
B737QC A320 QC (Pass) i Delta B737 Delta B737 A321 QC Delta Delta B737 1
(Pass.) (in %) (Pass) B737 (in %)
|Fuel Cost 0.857 0.793 i -0.064 -7.50% 0.742 -0.115 -13.41% |
lAirframe 0.598 0.610 0.012 2.04% 0.499 -0.098 -16.47%
maintenance
1 Engine maintenance 0.168 0.179 0.011 6.70% 0.218 0.051 30.15% 1
| Landing fees 0.482 0.533 0.051 10.66% 0.461 -0.020 -4.16% 1
|Handling fees 1.164 1.164 0.000 0.02% 1.126 -0.038 -3.28% 1
1 Navigation charges 0.729 0.713 -0.016 -2.23% 0.582 -0.147 -20.20% J
iTotal variable costs 3.997 3.992 -0.005 -0.14% 3.629 -0.368 -9.21% 1
I Fixed technical costs 0.864 0.861 -0.003 -0.37% 0.721 -0.143 -16.59% 1
| Capital costs 2.543 2.843 0.300 11.78% 2.553 0.010 0.38% 1
1 Aircraft 1.304 1.449 0.146 11.18% 1.302 -0.002 -0.15%
j depreciation
1 Spares 0.111 0.132 0.021 18.79% 0.118 0.007 6.28%
1 Depreciation
| Aircraft interest 1.040 1.157 0.117 11.20% 1.039 -0.001 -0.12% |
| Spares interest 0.089 0.105 0.017 18.97% 0.094 0.006 6.29% 1
|lnsurance 0.055 0.061 0.006 11.49% 0.055 0.000 0.06%
|Cockpit crew , 0.647 0.560 -0.088 -13.57% 0.430 -0.217 -33.52% 1
ICabin crew 0.491 0.425 -0.067 -13.57% 0.421 -0.070 -14.29% |
|Total fixed costs | 4.601 ; 4.749 0.148 3.22% 4.180 -0.421 -9.15% |
ITotal direct costs per 8.598 8.741 0.143 1.66% 7.809 -0.789 -9.18%
ISKO
DOC for the respective unconverted aircraft were shown to identify any
improvement or deterioration of cost differences between the different aircraft types. The
freight capacity was not taken into consideration in the DOC calculation, the passenger
Figure 12 shows only cash costs which were defined as crew costs, insurance
costs, handling, navigation, and landing fees, fuel costs, and maintenance costs. Capital
• Navigation
charges
° Handling fees
• Landing fees
• • Engine
maintenance
Figure 12. B737, A320, and A321 cash costs per SKO in normal and quick change-
passenger configuration.
costs and fixed maintenance costs were excluded. It was assumed that the variable
maintenance costs are cash costs. Cash operating costs for the B737 are 5.1 US cents per
SKO and 5.2 US cents per SKO for the QC version. Respective values are 4.9 and 5.0 for
the A320, and 4.4 and 4.5 for the A321. Therefore, if only cash costs are considered, the
A320 aircraft offers a potential advantage. The A321 is cheaper in both cases.
The respective costs for the cargo version are shown in Figure 13, Figure 14, and
Table 9. As a cargo aircraft, the A320 has 2.7% lower variable operating costs per TKO
but still 1.8% higher total DOC per TKO which is mainly caused by higher capital costs.
Since the capital costs of the passenger version were not affected significantly by the
conversion, it can be assumed that the high basic aircraft price causes the high capital
costs. In looking at cash costs only, values for the A320 QC and A321 QC are 36.3 US
•Navigation
charges
• H andling fees
'Landing fees
• Engine
maintenance
a
J2 •Airframe
maintenance
•Fuel Cost
'insurance
•Cockpit crew
'Fixed technical
costs
•Capital costs
Figure 13. B737 QC, A320 QC, and A321 QC DOC per TKO in cargo configuration.
48
•Navigation
charges
•Handling fees
•Landing fees
'Engine
maintenance
•Airframe
maintenance
• F u e l Cost
^Insurance
•Cockpit crew
Figure 14. B737 QC, A320 QC, and A231 QC cash costs per TKO in cargo configuration.
Table 9.-- B737 QC, A320 QC, A321 QC DOC per TKO in Cargo Configuration
(US Cents per TKO)
B737 QC A320 QC (Cargo) Delta B737 Delta B737 A321 QC Delta B737 Delta B737
(Cargo) (Cargo)
Total fixed costs 31.783 33.821 2.039 6.41% 31.452 -0.331 -1.04%
Total direct costs 64.057 65.232 1.175 1.83% 60.387 -3.670 -5.73%
49
cents per TKO and 33.0 respectively, compared to 37.7 of the B737. The cash cost
figures are more meaningful due to two reasons. First, the aircraft price is negotiable.
capital costs should not be considered since the aircraft is available anyway (if the airline
has the aircraft already in its fleet). Therefore, both aircraft offer a cost saving potential.
Two cases were analyzed. First, the effect of changes in fuel prices was analyzed.
This did not cause significant cost shifts. The DOC changed by less than 1% even if the
fuel price was doubled. This is due to the fact that the B737-300 is already a fuel efficient
The second case analyzed was zero conversion costs for the A320. In this case,
total DOC for the cargo version are reduced to 63.7 US cents per TKO which is 0.5% less
than the B737 value. Since the A320 is a bigger aircraft, the DOC per TKO should be
significantly lower than the B737 DOC due to economies of scale but that is not the case.
This indicates that the basic aircraft price is too high for the A320.
50
6.1 Overview
In the previous chapter, the A320/321 QC aircraft were assessed using a standard
method that does not consider the characteristics of its daily operation in A-Air. This
includes varying cargo volume over time, scheduling constraints, additional cost that were
not included in the standard DOC calculation, and other limitations that may affect the
optimum fleet mix and thus the decision whether or not to buy an additional QC type. For
example, although the A321 QC may have lower DOC as a cargo aircraft than the B737,
the lower passenger load factor during the aircraft positioning flight may eliminate any
cost advantage. Since an airline has to consider the network as a whole, it may prefer to
operate the B737 QC and lose some cargo due to insufficient capacity, because this is still
cheaper than acquiring and operating the larger A321 QC. Therefore, a model was
developed that takes more than only direct operating costs differences into consideration.
First, as mentioned above, an airline does not necessarily have to accommodate all
the cargo demand on a certain route. It may decide to operate a small aircraft and satisfy
only that part of the freight market that has a high enough yield to make a profit. In this
case, the airline looses some revenue due to insufficient capacity which is an opportunity
cost to the airline. As long as these opportunity costs do not outweigh the higher total
costs of operating a larger aircraft, the airline is better off using the smaller equipment.
51
Second, the operating costs of QC cargo aircraft may not be compared without
has higher DOC than an unconverted passenger aircraft. Depending upon the daily
utilization, the difference will impose additional costs to the QC cargo network. In this
context, it is often argued that the QC operation increases the total utilization of the
aircraft and therefore spreads the fixed costs (especially capital costs) over more flights.
This, in turn, would reduce the DOC of the aircraft and outweigh the penalties of the
conversion. But opinions on this aspect are inconsistent. Some argue that QC aircraft fly
less during the daytime, because route scheduling tries to avoid them (higher variable
cost). Therefore, fixed costs would not be reduced or spread over more hours. On the
other hand, an airline has to hold some spare aircraft. Of course, it will try to hold the
aircraft with the highest variable costs as spare capacity, in this case the QC aircraft. If
there would be no QC aircraft, the airline would have to use unconverted aircraft as spare
capacity. Therefore, there would still be some effect of reduced fixed costs. Since it was
not possible to determine which of the two arguments is true, the reduction of capital
costs was not considered in the model, but only higher direct operating costs due to the
Capital costs of the basic aircraft were excluded from the model. First, as
mentioned above, effects on capital costs are not yet analyzed in detail. Second, it can be
argued that the cargo operation with QC aircraft is a by-product. The basic aircraft are
available anyway and differences in capital costs should not be considered during cargo
operation. Therefore, only the capital costs of the conversion were included.
A fourth cost group to be included in total network cost considerations are the
costs of positioning flights at low load factors. Often, passenger service is offered on a
particular route only because it is necessary that the aircraft be available for night
operations at the destination. Therefore, a portion of the positioning flight costs has to be
Each additional conversion station causes investment in seat vans that is not
network, this is an additional cost above the normal handling charges that are not covered
by the direct operating costs. A larger aircraft may require additional conversion stations,
if the routes that were previously served by a one stop service with the smaller aircraft are
Finally, the structure of the schedule may require ferry flights, if additional aircraft
types are operated. Although the night schedule is only part of a whole schedule, it must
still be balanced with the same aircraft type arriving in the morning at a station as it was
schedule with a fleet mix that minimizes the above listed cost types. The three aircraft
available to this model are the B737, the A320, and the A321 as outlined in the previous
chapters. The model is an integer linear program that minimizes the objective function
rotations as main unknown variables. Feasible rotations are determined by the time
schedule of the flight plan. The principle is illustrated in Figure 15 using a small network
( 3 , 2 t 1
*1
B A C I)
4 ,— • ^ 5
>
6 ><
consisting of the stations A, B, C, D and the legs 1, 2, 3,4, 5, and 6. A rotation is defined
by its arrival and departure leg. For example, rotation 1-2 is the rotation that arrives from
leg 1 and departs to leg 2. It is feasible, if the departure time as stated in the schedule is at
least 30 minutes after the arrival time of the incoming leg (if 30 minutes is the minimum
transit time) and if it arrives/departs from the same station. Rotation 1-5, for example, is
not feasible, because an aircraft can not arrive at station C and depart from station A.
Rotation 6-1 is also not feasible, because the departure time of leg 1 is before the arrival
54
time of leg 6 (if, as assumed, the time schedule determines leg one to be the evening flight
Thus, rotations 1-2, 2-5, and 2-3 are examples of feasible rotations. It is also
feasible to origin at each station (unless explicitly excluded). That is, rotation 0-1 is
feasible. This states that the aircraft starts its nightly routine at station D. Same is true for
termination. Leg 1-0 is also feasible. In this case, the aircraft arrives from leg 1 and
terminates at station C. The costs of each leg as outlined below are allocated to each
rotation by its departure leg. For example, the DOC of leg 3 are allocated to rotation 2-3
In the following, chapters 6.2 and 6.3 explain the cost components of the objective
function and the rational of the constraints respectively. Chapter 6.4 will explain the
mathematical model and the notation which will also appear in the output of the computer
solution.
The objective function is a sum of all cost components over time discounted to
present value at any given discount rate. It is defined as the sum of variable cash
operating costs, costs of insufficient capacity (opportunity costs), additional costs daytime
operation, capital costs of the conversion, costs of positioning flights at low load factors,
55
conversion station costs, costs of ferry flights, and costs of idle aircraft. The method of
The variable cash operating costs were derived from the standard DOC calculation
as presented in chapter four. The costs included are fuel costs, maintenance costs, landing
fees, navigation and handling charges. Although some cost elements (such as landing
fees) may vary depending upon the routes flown, this fact was not taken into
consideration. It was assumed that the effects of lower or higher actual costs would
balance out for the network as a whole and would affect all three aircraft types by the
same amount. If, for example, the landing fees at one airport are higher than average, they
will be higher for all three aircraft types and not particularly high for one aircraft only.
Therefore, the variable cash operating costs are only a function of distance with a fixed
cost component.
The individual values of the DOC calculation for different stage lengths were
regressed with distance as independent variable and DOC per leg as dependent variable.
Landing and handling charges are independent from the stage distance and were treated as
fixed costs per leg. Their value depends upon the aircraft type (MTOW, max. payload).
Fuel and maintenance cost are a linear function of distance. Fuel costs would also vary
with the actual aircraft weight, however fuel consumption at max. payload similar to the
DOC calculation was assumed. Navigation charges are a nonlinear function of stage
56
distance. ATC fees decrease the further an aircraft leaves central Europe. With a model
similar to
or
ATC costs could be predicted by +- $10 and total DOC by +- $20 which represents
an error of about 1%. The adjusted R2 of the models were all larger than 0.9999 with F
Statistics between 200,000 and 400,000. The models were used as a basis for an Excel
spreadsheet where different input parameters such as fuel costs per liter, maintenance
costs, or changes in landing fees over time could be changed to analyze different scenarios
and to calculate DOC over time. This will be further presented in Chapter 6, "Input
Parameters".
Each route has a certain demand for overnight cargo service with a specific yield.
Depending upon this yield it is more or less important for an airline to satisfy all the
demand, or offer only limited capacity. The opportunity costs were determined by
multiplying the aircraft capacity with a target load factor and subtracting this value from
the demand on the particular route. If this value was positive, it was multiplied with the
opportunity costs per ton of not accommodated cargo demand. As long as it remained
negative (enough capacity), opportunity costs were zero. The opportunity costs can
Opp. Cost = (Demand Max. Payload * Target load factor) * Opp.Cost per ton
or 0 otherwise.
If the stage length of a particular route would exceed the optimum range of an
aircraft, the max. payload was adjusted accordingly. An expected cargo profile or an
expected (unconstrained) cargo growth rate establishes the value of the opportunity costs
over time. Different opportunity costs per ton may reflect the expected development of
the yield over time. Different maximum payloads of the different aircraft types allocate
opportunity costs for each leg for each aircraft. The opportunity costs were added to the
6.2.3 Additional Costs Daytime Operation and Capital Costs of the Conversion
The additional costs of daytime operation were estimated by subtracting the DOC
per block hour of the unconverted aircraft from the DOC of the QC passenger aircraft and
multiplying the difference with the daily utilization. This value per QC aircraft was
multiplied with the number of QC aircraft available during the planning period.
The capital costs were determined by subtracting the annual costs of the
unconverted aircraft from the annual capital costs of the QC aircraft. This value was then
divided by 365 to determine the daily costs. This was done because all costs should be
based on the same time horizon. If annual costs would be used, they would be
Cost of positioning flights were determined on a per leg basis similar to the
variable cash operating costs. Basis for the cost function were the data of the QC aircraft
in passenger configuration. It was assumed that the origin of the positioning flights would
be the main hub. Therefore, the stage distance equaled to the distance between the hub
and the conversion station. The costs of the positioning flights should decrease with
increasing passenger volume. In case the load factor reaches break-even load factor, there
should be no costs associated with the positioning flight. The costs of positioning flights
Costs of p-flights = 2 * DOC * (1- actual load factor/break even load factor)
In case the actual load factor exceeds break-even, costs of positioning flights are
set to zero, which means that there are no costs of positioning flights associated with this
route. In case of no passengers, costs of positioning flights are equal to total DOC on this
leg, which is about equal to the costs of a ferry flight. The DOC were multiplied by two
since each positioning flight consists of two legs (outbound and inbound). The costs of
the positioning flights were added to the total network costs in case an aircraft originates
from the station. Conversion station costs are based on the costs of the seat vans.
substantially increases the complexity of the model without significantly improving the
outcome. If scheduling a particular aircraft type (e.g. A320 QC) would require a ferry
flight, any rough estimate of the costs of this ferry flight will eliminate minor cost
advantages of the aircraft over other aircraft (e.g. B737 QC) and avoid the ferry flight by
scheduling the other aircraft, where the ferry flight is not required. If a ferry flight can not
be avoided with any aircraft type, the costs of the ferry flight are unimportant, because the
flight is necessary to accommodate the schedule. Therefore, a standard cost for all ferry
flights that reflects an average stage length of the network was utilized. Also, the model
normally imposes several other costs to a ferry flight (positioning flights may become
necessary, additional costs of a conversion station, etc.). Therefore, ferry flights are the
exception and can be covered with rough (conservative) estimates of the ferry flight costs.
Costs of idle aircraft are the costs of maintaining a fleet of aircraft that are
converted to QC aircraft but are not operated as cargo aircraft during the night. These
costs can be set as any large number, which means that the model minimizes the number
of aircraft that are not utilized, or they can be set equal to the additional costs daytime
operation plus capital costs of the conversion. Introducing costs of idle aircraft is
important, if the number of buy/sell transactions per period is limited and fluctuating
cargo volume might justify maintaining a larger than minimum fleet of different sized
aircraft.
61
These constraints can be grouped into two classes. First, there are those
constraints that are essential to generate a workable rotation plan. Second, there is a
group of optional constraints that are imposed by the airline's management. The first
group includes flight coverage, continuity of equipment, aircraft availability, and schedule
balance constraints. The second group is not exhaustive. One important constraint of this
group, however, is the limitation of buy/sell transactions over time. The shorter the
individual time periods become, the more important becomes this constraint, since
seasonal fluctuations in cargo demand may cause the model to change the equipment
Flight coverage refers to the necessity that each leg in the schedule has to be
covered once. Otherwise, the LP would assign zero aircraft to each leg, since the
necessity to cover the demand is not explicitly stated in other constraints but is included in
Continuity of equipment refers to the necessity that each flight has to begin and
end on the same aircraft type. This constraint is imposed by the mathematical formulation
Aircraft availability limits the number of aircraft that can be used to the number
that is available. This is either a number specified, or results from the objective function.
Schedule balance refers to the requirement that the schedule can be operated
continuously without accumulation of aircraft at a station. This means also that at each
conversion station the same aircraft type is converted to a cargo aircraft in the evening as
6.4.1 Notation
Xi j,k,t : Feasible rotation connecting from leg i to leg j on aircraft type k at time t.
time t.
Cdk,t : Additional costs daytime operation of one converted aircraft type k at time t.
Rsk,t : Maximum rate of aircraft of type k that may be sold during period t.
Rbk,t ' Maximum rate of aircraft of type k that may be bought during period t.
Kk,t : Minimum number of aircraft type k that have to be in the fleet at time t.
T f T
(1
>£ n£-
MIL Operating costs plus opportunity costs
(2)
k=\ i=0 ;=1
Additional costs daytime operation plus
(3) +]TaJU-(Cd*,f + CcJu)+ capital costs of the conversion
k=\
L M
Costs of positioning flights plus
(4) +X2 X °' , ' Jt ' /, ( C / 7 ''*' / + C ^'0' h conversion station costs
/=1 * = 1
5 M
Costs of ferry flights (Origination plus
(5) +^^C/i,^/-(05^,r-hr,^,r) + Termination shortage)
The first term (1) of the objective function states that the network costs are the
sum over the number of planning intervals of the planning horizon, discounted at a given
interest (discount) rate. If discounting is not desired, the rate may be set to zero.
The second term (2) expresses the sum of cash operating costs and opportunity
costs. Note that the costs are allocated to the departure leg of a feasible rotation.
Therefore, the costs of a i-0 rotation are zero since there is no departure leg. However,
with the constraint of continuity of equipment a i-0 rotation will require a 1-i rotation.
The 1-i rotation will have the costs of the i-leg assigned to it. Therefore, an i-0 rotation
implicitly gets costs assigned. With the constraint of schedule balance, the model will not
Term (3) refers to available aircraft in the fleet. In conjunction with the constraint
that the aircraft operated in the fleet (sum of all x0>1,k,t rotations) at any given time plus the
number of aircraft not operated during the same time period (yk,t) this part of the objective
function assures that ownership costs are associated with each aircraft in the fleet.
65
Term (4) implies that for each originating flight (0-i rotation) a positioning flight
is required and the conversion station costs incur. In theory, it might be possible that
there are several flights originating at a station with economies of scale associated.
the same station will be rarely found. Therefore, conversion station costs were allocated
An origination shortage (5) occurs if there are more arrivals than departures at a
station. In this case, the excess of arriving flights has to leave the station as a ferry flight
of origination shortage that is, there are less arrivals than departures or a shortage of
Term (6) imposes an additional cost penalty for idle aircraft. Normally, the cost of
idle aircraft are already included in term (3). Idle aircraft, however, should be avoided
and therefore an additional cost penalty was imposed. If idle aircraft are not considered
an additional penalty, term (6) may be omitted and subsequently constraint (C) has to be
changed in that y is omitted from the formula and the equal sign is replaced by a less than
or equal sign.
66
6.4.3 Constraints
The first constraint states that each leg has to be covered once by exactly one
aircraft type.
M L
The constraint states that the sum of all rotations for all aircraft types that depart
into leg j (that is depart into leg 1, 2, ,L) from any arriving leg i has to be equal to one
for all planning intervals. Therefore, the LP is forced to pick one, and exactly one aircraft
type for each leg of the network during each interval. However, it may chose different
(B) states that if a certain aircraft type (e.g. a B737) was picked to depart into leg 1
(e.g. 3) at time t (e.g. 1), there must be a departure into any leg i with (in this example) a
B737 arriving from leg 3. Or, in other words, the same aircraft type has to depart into,
and arrive from the same leg. This must be true for all rotations and all aircraft types at
all times. Of course, the aircraft may also justflythis single leg (0-1 and 1-0) in which
67
case the model may have to assign a ferry flight, if the schedule is not balanced any more.
Constraint three (C) limits the number of aircraft used in the schedule:
(C) X * 0 ' * < + yk < = ak< for all k = 1,...,M and t = 1,....,T
1=1
If ak,i is not externally given as the number of already existing QC aircraft in the
fleet (Kk,t), the model plans the number of aircraft required without consideration of
already existing aircraft. The number of aircraft in operation (sum of all x0,i,k,t) plus the
sum of all aircraft not used during the period (yk,t) must be equal to the number available
as stated in the objective function (ak,t). This constraint has only an effect on the optimum
solution, if the number of buy-sell transactions per period is limited (see constraint (E)).
That is, the sum of all flights on a certain aircraft type k at time t leaving from a
station s, plus the number of excess arrivals, must be equal to the sum of all aircraft of
68
type k arriving at station s, plus the number of excess departures at time t For example,
if, at time 1, there are four B737 flights departing from station A and six B737 flights
arriving at the same station A, then OA, 13737 must be equal to two with the subsequent
The fifth constraint is optional and limits the number of buy sell transactions per
period It may appear in different forms dependent upon the airline's situation In the
following, it was assumed that the airline keeps converted aircraft for at least seven (=TS)
years before they may be sold After seven years, no more than two aircraft per period
may be sold of each aircraft type Initial number of aircraft is five (=K7^71) B737 QC that
each period l
1
Tins has only then a limiting function if t ^ U Otherwise, (El) limits the number of aircraft that ma>
be sold to zero
69
A similar constraint can be build to limit the number of aircraft purchases per time
period
However, limiting the number of aircraft that may be bought may result in an
infeasible solution, if not enough aircraft to cover the schedule can be purchased The sell
and buy rates can be chosen by the airline and depend upon the airline's financial situation,
the duration of one planning interval, and the manufacturer's ability to convert aircraft In
Finally, all variables in the model have to be positive integer numbers with all x,0,u
being binary numbers However, in most cases, due to the construction of the model, the
7 INPUT PARAMETERS
The following chapter outlines the input parameters as they are used in the model.
A complete presentation, however, will not be possible due to the large number of
individual values that enter it. The final data set consists of about 42,000 observations
containing about 10,000 nonzero values for variables of the objective function. Therefore,
All values are calculated using Excel spreadsheets. The basic data used in the
economic comparison served as database. The values were then exported into SAS
Software where they were sorted, modified and combined to the sparsedata format that is
a main hub (A) and 19 stations (B-T) as illustrated in Figure 16. Some services (e.g. D-A
or K-A) are one-stop services, which means that they have an intermediate stop at another
station before arriving at the hub (e.g. service D-A has an intermediate stop at station C).
All incoming legs to station (A) have to connect to the outgoing legs. Except for the
29
Refer to SAS Institute, SAS/OR User's Guide, Version 6 (Cary: SAS Institute, 1989) chapter 7 for
details about using SAS/OR for solving LP problems.
71
c 1 38 37
>
B S T=2 T
2
i r 35 ,. 36
k
**
', 4 3 34
c *»
- "?
D C R
5
> 6
53 „ 33
> >C - —
c 8 c 7 32 y 31 5
t " j?
>
F E P Q
9 y 10
> c 29 . 30
J
< n 28 27
G A N O
12
i cK 25 ,, 26
**
c 13 24 >
H T=3 M
14 23
*
>
7> c
t. "••"
c 15 22 j
I T=2 L
16 21
> c
k--1--'1-1 •"••••
18 17
€ c
%
K J T=3
19 20 y
J
hub. Service from station (F) is independent and connects only stations (F) and (E).
Service to stations L, S, and T starts at time 2 (second year), and to stations H, J, and K at
time 3 (third year). All legs were assigned numbers that will appear in the output of the
LP. For example, rotations 2-1 will have leg 2 (inbound from station (B) to station (A))
as arrival leg and leg 1 (outbound from station (A) to station (B)) as departure leg. The
number of feasible rotations in the model amounts to 9,027 for all time periods or about
The network includes all routes where cargo service may be offered in the future
and QC aircraft may be scheduled. It has to be noted that the model does not consider
alternative aircraft such as pure freighters. It determines only how many QC aircraft of
each type should be operated, in case the airline decides to acquire QC aircraft. That is, it
determines the competitiveness of the Airbus QC aircraft compared to the B737 QC. The
airline may still decide to operate pure freighter aircraft, charter capacity, or not to serve a
route at all. Therefore, the analysis provides an upper limit for potential demand of
A320/321QC aircraft.
The time horizon is twelve years. This gives a good long-term picture of the fleet
planning requirements with the assumed growth rates of passenger and cargo volume. If
purchase and sell restrictions are omitted, it also indicates when the Airbus aircraft
become competitive.
Network input parameters are illustrated in Table 10. Distances range from 150
km to 2,000 km, which is about the optimum range of the B737 QC. Longer distances are
73
Table 10.-- Leg Designators, Distances, Initial Demand and Opportunity Costs of the
Network.
infeasible for QC operation, because the departure time of the associated passenger flights
has to be too early in the evening to accommodate the departure time requirements of the
of demand is not considered. Initial screening of available cargo data showed that
demand does notfluctuateconsiderably. Also, fluctuations within the year balance out
since time intervals of one year are chosen. The target revenue cargo load factor was set
to 80%. This means that if the actual demand in tons on a particular route exceeds 80%
of the payload of the aircraft, opportunity costs will be assigned to this aircraft on this
route. The opportunity costs per ton remain constant over time.
Initial stock of aircraft isfiveB737 QC. They may be replaced at the earliest after
five years. However, the LP was run twice, once without the sale constraint. This is to
separate potential weaknesses of the Airbus aircraft from purchase restrictions of the
airline. Expected passenger demand is shown in Table 11. This demand is expected to
grow by 3% per time period. Revenue passenger break-even load factor is assumed to be
60%.
Boeing Commercial Airplane Group, 1994 World Air Cargo Forecast. (Seattle: July 1994) cargo
forecast for Europe.
75
Table 11.-- Initial Average Passenger Demand to Each Station in the Network.
seen in Figure 17. The chart was created by adding all individual values of each cost
group for all time intervals and all aircraft types and then showing the total amount of
31
Average demand is defined as the mean of passengers on the evening flight and passengers on the
early morning flight.
76
each cost group relative to the sum of all cost components. Therefore, the chart shows
how the individual cost components are represented in the model. A high percentage
value indicates that minimizing the respective cost component takes a high priority in the
DOC make up for only 22% of the sum of all nonzero cost components in the
objective function which is about the same weight as the opportunity costs. Since
opportunity costs are a function of insufficient capacity, this shows that the capacity
aspect of the network will become increasingly important. Idle cost penalty has about the
same value (12%) as the sum of capital costs plus additional costs of daytime operation.
This means that having an idle aircraft in the fleet imposes capital costs and daytime
77
operating costs twice to the aircraft. The ferry flight penalty has about the same amount as
the DOC which means that costs of a ferry flight are about as high as the average costs of
a leg in the network. Positioning flight costs are about half as important as the DOC but
still make up a considerable amount of the components that can not be avoided in
operating the network (costs of idle aircraft and ferry flight costs can be avoided). Costs
of a conversion station are of minor importance, since only the costs of the seat vans are
Respective values similar to Figure 17 for each individual aircraft type are shown
in Table 12. The weight of direct operating costs is about the same for all aircraft.
Opportunity costs are weighed highest for the B737 which results from the fact that it is
the smallest aircraft and demand exceeds 80% capacity on several routes over time. All
78
other cost components are weighed higher for the Airbus aircraft This indicates that the
higher capacity (lower opportunity costs) results in higher aircraft related costs
The absolute values are given in Table 13 Note that the absolute values are
decreasing over time since they are discounted to year one at 9% Additionally, the
number of legs served, and different growth rates of parts of the components as given in
Table 14 affect the average values Graphically, the figures are presented in Figure 18 It
can be seen that the difference of total cost between the B737 and the Airbus aircraft
The absolute direct operating cost advantage of the B737 is more than outweighed
by the lower opportunity costs of the Airbus aircraft (lower two pieces of the bar chart)
With the higher daytime operating costs (third piece), all three aircraft are about even
Capital costs of the conversion (fourth piece) are similar for all aircraft and do not change
the cost structure The cost disadvantage of the Airbus types in the positioning flight costs
causes an absolute cost disadvantage in the earlier periods, until passenger demand is
assumed to pick up Conversion station costs are negligible Ferry flight costs and costs
of idle aircraft are of minor importance, since, as mentioned above, they can be avoided
79
Table 13.- Average Costs per Cost Component, Aircraft Type, and Time Period.
1 Time Aircraft Average Average Daytime Capital Average Conversio Cost Costs of I
DOC Opportunit costs costs Costs of n Station penalty of Idle
yCost Positioning Costs ferry Aircraft I
Flights flights
1 8737 $4,397 $4,167 $1,100 $1*001 $1,648 $161 $4,000 $2,600 I
1 A320 $4,361 $2,642 $1,526 $1,307 $2,331 $161 $4,700 $3,000
1 A321 $5,438 $1,045 $2,676 $1,481 $3,517 $161 $5,300 $4,000
2 * B737 $4,200 $3,992 $1,029 $918 $1,706 $148 $3,853 $2,294
2 A320 $4,648 $2,510 $1,428 $1,199 $2,380 $148 $4,528 $2,752
2 A321 $5,199 $1,062 $2,505 $1,359 $3,525 $148 $5,106 $3,670
3 B737 $3361 Man $963 $842 $1,424 $136 $3^12 $2,104
3 A320 $4^88 $2,423 $1,337 $1,100 $2,077 $136 $4^61 $2,525
3 A321 $4,909 $1,080 $2,344 $1,247 $3,204 $136 $4^18 $3,367
4 B737 $3,750 $4,013 $901 $773 $1,272 $124 $3,576 $1,930
4 A320 $4,157 $2,689 $1,251 $1,009 $1,892 $124 $4,201 $2,317
4 A321 $4,650 $1,313 $2,193 $1,144 $2,964 $124 $4,738 $3,089
5 I B737 $3J552 $4,266 $844 $709 $1,134 .: $114;.,;... $1,771
5 A320 $3£38 $2,968 $1,171 $926 $1716 $114 $4^47 $2126
5 A321 $4,405 $1,559 $2,052 $1,049 $2,739 $114 $4,564 $2,834
6 B737 $3,364 $4,550 $789 $650 $1,013 $105 $3,318 $1,625
6 A320 $3,732 $3,285 $1,095 $849 $1,555 $105 $3,899 $1,950
6 A321 $4,174 $1,822 $1,921 $963 $2,528 $105 $4,396 $2,600
\ :
7: % B737": r
$3,187 $4,314 $739 ,;;;$9osT:; $96;,:, $3*196 $1,491
7 A320 $3£38 $3,555 $1,025 $779 $1,403 $96 $3J56 $1,789
7 A321 $3,956 $2,089 $1,797 $883 $2£30 $96 $4,236 $2,385
8 B737 $3,019 $5,033 $691 $547 $822 $88 $3,079 $1,368
8 A320 $3,352 $3,784 $959 $715 $1,261 $88 $3,618 $1,641
8 A321 $3,750 $2,318 $1,682 $810 $2,145 $88 $4,080 $2,188
9 B737 $2,861 $5,211 $647 $502 $748 $81 $2,966 $1,255
9 A32G $3,178 $4,033 $898 $656 $1,128 $81 $3^86 $1,606
9 A321 $3,556 $2,558 $1,674 $743 $1,971 $81 $3^30 $2,007
10 B737 $2,712 $5,369 $605 $461 $680 $74 $2,857 $1,151
10 A320 $3,013 $4,272 $840 $602 $1,007 $74 $3,357 $1,381
10 A321 $3,371 $2,831 $1,473 $682 $1,808 $74 $3,786 $1,842
/ $626 $68 $2,752 -^ $1,056
B737 $2,571 $5,679 $566 '"'$423'"~
11
A320 $2,857 $4,470 $786 $662 $900 $68 $3,234 $1,267
11
11 A321 $3,197 $3,064 $1,378 $626 $1,655 $68 $3^47 $1,690
12 B737
!
' $2,438* $5,747 $530 $388 $578 $62 $2,651 $969
12 A320 $2,710 $4,632 $736 $507 $808 $62 $3,115 $1,163
12 A321 $3,032 $3,260 $1,290 $574 $1,512 $62 $3,513 $1,550
80
$20,000
S 18.000
$16,000
J ^t
$14,000
$12,000
$4,000
i *l;|?' 1 fl ": 1 ill *:l ill 1 ji II t M If! IIffl1 If: 1 if 1 II II 1 I II I I I I I ! |j I
• ": ** " *• * ii m \k u " i & w ;i i * || i ti | il ft
$2,000
$0
• AverogB DOC • Average Opportunity Cos f D o y f i r r e costs QCcpltdcosts • Average Costs of Positioning Flights OConvefs Ion StoHcn Costs • Cos tpendty of ferry flights • Costs of IdeAircraft]
Relative average values of the objective function cost components are shown in
Figure 19. It can be seen that the weight of the DOC remains about constant over time
I MllllllllllllllllllllllllllII
II 111 III i l l 11:11 ill 11111111:1
iiiiiiiiiiiiiiiiimiiiiiiiiiiiiiii
0%
,
i il mini
1 1 2 2 2 3 3 3 4 4 4 S 5 S 6 6 6 7 7 7
i i i iiiiimi
B737 A320 A321 6737 A320 A321 B737 A320 A321 B737 A320 A33I B737 A320 A321 B737 A320 AJ21 B737 A320 A321 B737 A320 A321 B737 A320 A321 B737 A320 A E 1 B737 A320 A321 B737 A320 AS21
8 8 8 9 9 9 l O B T O II 11 11 B B C
| • A v e r c g e D O C P Avercge Opportunity C a t • Doytlrrecosls aCcpttdccsis •Average Costs of Pes rrtanhg Flights DConverslcnStcitcnCosts •Ccstpendtycffenvfllohls BGcsftcfldeAircrcft
between 19% and 24%. The weight of the opportunity costs, however, increases from
about 22% to 43 % in case of the B737 QC, 12% to 33% in case of the A320 QC, and 4%
to 22 % in case of the A321 QC. The weight of the positioning flights decreases over
time which is a result of the assumed growth in passenger demand. The weight of the
All input parameters that determine the components of the network costs were
calculated for the first year and then recalculated for subsequent years assuming different
growth rates for each parameter A summary of estimated growth rates is given in Table
14
DOC for the first time period were calculated using the formulas
of the LP are included in Appendices B and C. The computations were done on a PC with
16 MB RAM and an Intel Pentium P90 processor board. Computation time ranged
between one and two hours, depending upon the number of constraints and the desired
output data sets. The LP procedure employs a two-phased revised simplex method.32
that begin with an "A" followed by four digits stand for the number of suggested aircraft
in the fleet at a specified time (first digit) for a specific aircraft (last three digits).
Designators beginning with an "X" specify a feasible rotation. The first group of digits
indicates the inbound and the outbound leg. The following group of letters indicates the
departure and arrival station. The last group of digits specifies time and aircraft type
similar to "A"-variables.
Section 8.1 presents the recommended fleet mix for the network with and without
the sales constraint as outlined in Chapter 5. In section 8.2 the aircraft rotation schedule
as derived from the SAS printout will be discussed. Section 8.3 contains a cost analysis
of the cost components in the optimum solution and further analysis of the
32
See SAS Institute, SAS/OR User's Guide, Version 6 (Cary: SAS Institute, 1989), 229 for details.
8.1 Fleet Mix
The suggested fleet mix is shown in Table 15. It can be seen that the sales
constraint is of importance to the results of the LP at the earlier time periods during the
expansion of the network. Initial recommended fleet size is ten aircraft comprising of
four B737, two A320, and four A321. If more than five B737 have to be in thefleet,one
A320 is traded against the fifth B737. With the expansion of the network in year two,
three aircraft are added to the fleet, which gives a similar fleet mix for both versions of
the LP. In year three, when the network is expanded further, the fleet is expanded by
another aircraft. In case an aircraft may not be sold before seven years, the model
85
suggests to acquire a larger A321 instead of the A320. In year four, both versions of the
model suggest the same fleet mix of six B737, three A320, and five A321 since the
version without the sales constraint has traded one A320 of period three against an A321.
In year five, another A320 is traded against an A321. If this is not allowed, the network is
covered with a similar fleet mix as in year four, however, with insufficient payload
capacity for this period. As soon as older B737 aircraft may be sold after period six,
Boeing aircraft are traded against the larger A321 which becomes the major aircraft type
in the fleet.
Suggested aircraft rotation plans without and with sales constraint are shown in
Table 16 and Table 17 respectively. The tables were derived from the SAS printouts
included in Appendix 2. For example, aircraft rotation 2-1 means that one aircraft flies
legs two and one during one night, which comprises of the feasible rotations 0-2, 2-1, and
1-0. For easier analysis, the original suggested rotation plan was slightly modified
without any impact on the total network costs of the optimal solution by forming closed
aircraft rotations. If, for example, the LP suggested to operate one A321 on legs 2-3-4
and one on legs 5-6-1 during the same time period, the result was modified in that one
Table 16.- Aircraft Rotation Schedule without Sales Constraint
A321 operates on legs 2-1 and the other one on legs 5-6-3-4. This has no effect on total
costs because feasible rotation 2-1 has the same costs assigned as feasible rotation 6-1.
Therefore, since the model is indifferent between picking feasible rotation 2-1 or 6-1 (all
constraints are met by both rotations versions), it was simply by chance that the LP did
not form closed aircraft cycles as it did in some cases (e.g. 0-10, 10-7, 7-0, time 2,
B737).33
Among the four rotations that should be served by an A321 from the beginning,
two are already served by the airline (26-25-28-27 and 23-24) using B737 equipment.
The other two routes are among the ones with the highest probability to be included in the
QC network (2-1 and 5-6-3-4). This means that the current equipment is getting too small
and should be replaced by a larger aircraft. Among the routes that should be served by an
A320, only one rotation (16-15, currently not served) should be flown with an A320
regardless of a sales constraint. The other rotation (33-34, currently served by a B737)
should be flown with an A320 after one year, as soon as the Boeing aircraft can be
33
Per definition, there is no cost advantage if not only the same aircraft type, but even the same aircraft
arrives m the morning at the same station as it departed the night before In fact, having closed aircraft
cycles in this case has no effect on costs.
89
recommended. Initially, DOC, opportunity costs, and the cost of positioning flights are
shown. This is followed by the equivalent data of the Airbus aircraft on the same legs,
including the relative difference to the B737 QC. Capital costs are not included, since
they are independent from a particular leg. Costs of conversion stations are included in
the positioning flight costs, since they are of minor importance to the optimum solution.
All cost data are derived from the LP with the sales constraint. The difference to the
solution without sales constraint is negligible and therefore omitted. The structure of
Appendices E and F is similar to Appendix D, however the A320 QC and the A321 QC
Figure 20 illustrates the weight of the different cost components in the optimum
solution. Idle and ferry costs are zero percent, since they were avoided at any period. The
weight of direct operating costs has increased to 47%, while opportunity costs have
decreased relative to the direct operating costs (in the input data both components have
Idle costs
Ownership
co,,s Fe
Posltloning " y penalty
flight costs 10%
11%
about the same weight). This indicates that on the average larger equipment with higher
DOC but lower opportunity costs is suggested by the LP. Since the LP could have
scheduled smaller aircraft at lower DOC, if this would have reduced total network costs
but did not do so, it can be concluded that the additional costs of a larger aircraft (A321
QC) in the given network are justified in the light of total network cost minimization. Or,
in other words, the marginal benefit of larger equipment (lower opportunity costs) is
higher than its marginal costs (higher direct operating, ownership, and positioning flight
costs). Consequently, the B737 QC is substituted against the A321 QC with increasing
demand over time, although the A321 QC has about 25% higher direct operating costs
Appendix 6). The LP suggests to operate an A321 QC on the rotation at all times. Even
91
in year one, with a relatively low passenger load factor (65 passengers) on a relatively
long positioning flight leg, the high opportunity costs due to high (even unbalanced) cargo
demand costs more than outweigh the additional costs of the A321 QC. Total B737 costs
of the rotation are about $84,000 higher than the respective A321 costs. This is
The A320 QC is represented with only few aircraft in the suggested fleet mix at a
decreasing tendency. This indicates that the range of route characteristics (cargo and
passenger demand, stage length, etc.) where scheduling an A320 QC offers cost
already advantageous to switch to the larger A321 QC. The time periods where the A320
the network where the A320 QC would offer a payload advantage due to range problems
of the other two aircraft. This means that the range advantage is not valued on any route.
However, with the scheduling problem of passenger flights on long positioning flights as
mentioned in a previous chapter, routes beyond 2,000 km will be the exception for a QC
network.
Table 18 gives an overview of the cost components for each aircraft type. The
relative weight of each cost variable indicates its portion of the total aircraft related costs.
The A321 QC opportunity costs have the highest weight compared tot the other two
aircraft, although it is the largest aircraft. This apparent contradiction is due to the fact
92
Table 18.- Aggregated Costs per Aircraft Type and Cost Component.
that the A321 QC can not be substituted by a larger aircraft if demand exceeds its capacity
as it is the case with the other two aircraft. In combination with the fact that the weight of
the total opportunity costs has decreased compared to the weight in the input parameters,
the fact of the high relative opportunity costs of the A321 QC is no contradiction.
Mean values in Table 18 indicate average annual costs of the respective aircraft
types. That is, average DOC of a B737 QC are $3,160, of an A320 QC $4,037, and of an
93
A321 QC $4,235 per leg. The mean of ownership costs indicates the average amount that
Assuming that the Airbus aircraft can be offered as technically specified, both
aircraft offer advantages and, mainly in case of the A321 QC, potential cost savings for A-
Air. The technical layout with the aft main cargo door eliminates one main QC specific
disadvantage which is the reduced comfort in the area of first or business class
passengers. Additionally, the threat of engine damage during loading and unloading is
reduced and access to the main entrance door IL is not disturbed by the cargo operation.
Currently, two of the five existing A-Air routes would support operation of an
A321 QC. Two of the potential new routes, one with the highest probability of being
realized, supplement potential short term demand to a maximum of four A321 QC within
the airline's fleet. Medium range demand of the A321 QC amounts to four to seven
aircraft, depending upon the rate of the network expansion. In the long run, increasing
passenger and especially cargo volumes may increase the number of required A321 QC by
another three to four aircraft. This research does not consider whether A-Air has the
financial resources to add to its QC fleet at this time. Also, this research does not
consider other alternative ways of serving A-Air's cargo market demand and routes which
A-Air's potential demand for an A320 QC is limited to less than three aircraft.
However, if the A321 QC cannot meet its technical specifications, potential A321 QC
routes may also be served by an A320 QC. Otherwise, the A320 QC is suggested only as
95
an intermediate aircraft type. The structure of the network with relatively short legs
within the optimum range of the other two aircraft does not provide a payload-range
advantage for the A320 QC. However, the A320 may represent a viable pure freighter
aircraft for medium range operations, where it can benefit from its longer range which
Abara, Jeph. "Applying Integer Linear Programming to the Fleet Assignment Problem."
Interfaces 19 (July/August 1989): 20-28.
Airbus Industrie. Performance Doc. P2210 Rev. 2, June 93 and Performance Doc.
P21131Rev.LMav92.
Borchard, Walter. A320 Feasibility Study. Deutsche Aerospace Airbus, February 1994.
Hammer, Robert H. "Fleet and Airplane Acquisition Planning of Regional Airlines." M.S.
Thesis, Massachusetts Institute of Technology, 1983.
Kirby, D. "Is Your Fleet the Right Size?" Operations Research Quarterly 10 (1959):
252; quoted in Christopher Colin New. "Transport Fleet Planning For Multi-Period
Operations." Operations Research Quarterly 26 (1975).
Sprenger, Wilfried and Karl Kwik. A320 OC Cabin Layout. Deutsche Aerospace, March
1994.
SAS Institute. SAS/OR User's Guide. Version 6. Cary: SAS Insitute, 1989.
Schick, GJ and JW Stroup. "Experience with a Multi-Year Fleet Planning Model." The
International Journal of Management Science 9 (1981): 389-96.
Table 19.- Summary of Input Parameters for the DOC per SKO and TKO Calculation.
Units B737-300 B737 QC B737 QC A320-200 A320 QC A320 QC A321-100 A321 QC A321 QC
Pass (cargo) (Pass) (Cargo) (Pass) (Cargo)
Stage length NM 500 500 500 500 500 500 500 500 500
Stage length KM 927 927 927 927 927 927 927 927 927
Payload kg 14,626 13,563 15,645 16,300 15,509 17,900 19,250 18,634 21,750
Available Seats 123 121 0 144 140 0 182 182 0
Available freight kg 2,572 1,705 15,645 2,188 1,789 17,900 1,414 798 21,750
Flight time hr 1.38 1.38 1.38 1.3 1.3 1.3 1.35 1.35 1.35
Block time hr 1.63 1.63 1.63 1.55 1.55 1.55 1.6 1.6 1.6
Block fuel kg 3,460 3,532 3,532 3,620 3,780 3,780 4,100 4,600 4,600
Annual Utilization:
Number of flights 1,920 1,920 1,920 1,920 1,920 1,920 1,920 1,920 1,920
Distance 1,000 1,779 1,779 1,779 1,779 1,779 1,779 1,779 1,779 1,779
km
Flight Hours hr 2,650 2,650 2,650 2,496 2,496 2,496 2,592 2,592 2,592
Block hours hr 3,130 3,130 3,130 2,976 2,976 2,976 3,072 3,072 3,072
Block fuel 1,000 6,643 6,781 6,781 6,950 7,258 7,258 7,872 8,832 8,832
kg
Ton-km offered 1,000 26,018 24,127 27,831 28,996 27,589 31,842 34,243 33,148 38,691
TKO
Seat-km offered 1,000 218,802 215,244 0 256,159 249,043 0 323,756 323,756 0
SKO
Table 20.-- Annual Direct Operating Costs
(in million USD)
Fixed technical costs 1.860 1.860 1.860 2.144 2.144 2.144 2.334 2.334 2.334
Capital costs 5.109 5.474 5.474 6.603 7.080 7.080 7.724 8.265 8.265
Aircraft depreciation 2.613 2.806 2.806 3.357 3.609 3.609 3.929 4.214 4.214
Spares Depreciation 0.229 0.239 0.239 0.315 0.328 0.328 0.367 0.382 0.382
Aircraft interest 2.085 2.239 2.239 2.679 2.881 2.881 3.135 3.364 3.364
Spares interest 0.182 0.191 0.191 0.252 0.262 0.262 0.293 0.305 0.305
Insurance 0.110 0.118 0.118 0.141 0.152 0.152 0.165 0.177 0.177
Cockpit crew 1.394 1.394 1.394 1.394 1.394 1.394 1.394 1.394 1.394
Cabin crew 1.058 1.058 0.000 1.058 1.058 0.000 1.364 1.364 0.000
Total fixed costs 9.530 9.903 8.845 11.339 11.827 10.769 12.979 13.532 12.169
Total direct costs 18.201 18.507 17.827 21.209 21.768 20.771 24.387 25.281 23.364
101
PROBLEM SUMMARY
Non-negative 10308
Total 10308
EQ 2466
Objective 1
Total 2467
SOLUTION SUMMARY
Terminated Successfully
n
OUjeeUve Jol i d ^ v c ,yp<_ uv^«_, . ^ ^ ^ ^ ^w_ _ ~ . d ^ z~>
row
bound cost
°r^' — R:J~
bound cost
x
- ' ^~ 'r ,r>r
* w
->lim iinn^r honnH Prirp Reduced
bound cost
PROBLEM SUMMARY
Total 10342
EQ 2466
GE 34
Objective 1
Total 2501
SOLUTION SUMMARY
Terminated Successfully
bound cost
row
bound cost
nhf^i-w^ VjqripM <-> Tvn<» | nwpr Value Unnpr bound Price Reduced
row bound cost
1 $3,597 $0 $0 $3,597
2 $3,404 $0 $0 $3,404
3 $3,222 $0 $0 $3,222
4 $3,049 $0 $0 $3,049
5 $2,887 $0 $0 $2,887
6 $2,733 $0 $0 $2,733
7 $2,588 $0 $0 $2,588
8 $2,452 $0 $0 $2,452
9 $2,322 $0 $0 $2,322
10 $2,200 $40 $0 $2,240
11 $2,084 $291 $0 $2,375
12 $1,975 $516 $0 $2,491
10
$30,494 $32,507 $1,932 $64,933
11 1 $4,109 $0 $0 $4,109
2 $3,889 $0 $0 $3,889
3 $3,681 $0 $0 $3,681
4 $3,485 $0 $0 $3,485
5 $3,299 $0 $0 $3,299
6 $3,124 $355 $0 $3,479
13 3 $4,229 $0 $0 $4,229
4 $4,004 $0 $0 $4,004
5 $3,790 $0 $0 $3,790
6 $3,589 $0 $0 $3,589
7 $3,399 $0 $0 $3,399
8 $3,220 $0 $0 $3,220
9 $3,051 $0 $0 $3,051
13 $25,282 $0 $0 $25,282
17 3 $3,738 $0
$0 $0 $3,738
4 $3,538 $0 $0 $3,538
5 $3,350 $0 $0 $3,350
6 $3,172 $0 $0 $3,172
136
7 $3,005 $0 $0 $3,005
8 $2,846 $0 $0 $2,846
9 $2,696 $0 $0 $2,696
10 $2,555 $53 $0 $2,608
11 $2,421 $388 $0 $2,809
3 $3,667 $0 $0 $3,667
4 $3,471 $0 $0 $3,471
5 $3,286 $0 $0 $3,286
6 $3,112 $0 $0 $3,112
7 $2,947 $0 $0 $2,947
8 $2,792 $0 $0 $2,792
9 $2,645 $0 $0 $2,645
10 $2,506 $0 $0 $2,506
11 $2,375 $0 $0 $2,375
$26,801 $0 $0 $26,801
3 $3,738 $0 $0 $3,738
4 $3,538 $0 $0 $3,538
5 $3,350 $0 $0 $3,350
6 $3,172 $355 $0 $3,527
7 $3,005 $801 $0 $3,806
8 $2,846 $1,202 $0 $4,048
9 $2,696 $1,560 $0 $4,256
10 $2,555 $1,879 $0 $4,434
11 $2,421 $2,163 $0 $4,584
2 $5,363 $0 $0 $5,363
3 $5,075 $0 $0 $5,075
4 $4,803 $0 $0 $4,803
5 $4,547 $0 $0 $4,547
$19,788 $0 $0 $19,788
=========== =============== =============== ===========
9 $2,590 $0 $766 $3,356
10 $2,454 $0 $591 $3,045
11 $2,326 $0 $428 $2,754
12 $2,204 $0 $276 $2,480
9 $2,590 $0 $0 $2,590
10 $2,454 $0 $0 $2,454
11 $2,326 $0 $0 $2,326
12 $2,204 $0 $0 $2,204
$9,574 $0 $0 $9,574
$4,272 $0 $0 $4,272
$4,042 $585 $0 $4,627
1 $3,639 $0 $0 $3,639
2 $3,444 $0 $0 $3,444
$7,083 $0 $0 $7,083
1 $4,272 $0 $0 $4,272
2 $4,042 $585 $0 $4,627
37 2 $3,773 $0 $0 $3,773
37 $3,773 $0 $0 $3,773
COST COMPONENT ANALYSIS
Absolut Values and Relative Differences to the Solution Aircraft Type
-Values in Brackets are Negative-
Leg TIME DOC 320 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
B737 A320 B737 A320 B737 A320 B737
$36,436 $0 $0 $36,436
Leg TIME DOC 320 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
B737 A320 B737 A320 B737 A320 B737
11 $24,058 $0 $0 $24,058
=== =========
13 3 $4,695 11% $0 $0 $4,695 11%
4 $4,447 11% $0 $0 $4,447 11%
5 $4,213 11% $0 $0 $4,213 11%
6 $3,991 11% $0 $0 $3,991 11%
7 $3,782 11% $0 $0 $3,782 11%
8 $3,585 11% $0 $0 $3,585 11%
9 $3,398 11% $0 $0 $3,398 11%
$0 $0 $28,111
13 $28,111
=== —————————
3 $4,695 11% $0 . $2,876 81% $7,571 30%
14
4 $4,447 11% $0 . $2,526 94% $6,973 31%
COST COMPONENT ANALYSIS
Absolut Values and Relative Differences to the Solution Aircraft Type
-Values in Brackets are Negative-
Leg riME DOC 320 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
B737 A320 B737 A320 B737 A320 B737
17 $30,501 $0 $0 $30,501
18 $29,938 $0 $0 $29,938
Leg TIME DOC 320 Diff.DOC Opp.Cost Diff.Opp Pos Cost Diff.Pos Total Diff.Tot
B737 A320 B737 A320 B737 A320 B737
22 $21,871 $0 $0 $21,871
— ========-
26 9 $2,900 12% $0 $1,612 110% $4,512 34%
10 $2,750 12% $0 $1,392 136% $4,142 36%
11 $2,608 12% $0 $1,188 178% $3,796 38%
12 $2,473 12% $0 $997 261% $3,470 40%
27 $10,731 $0 $0 $10,731
:== ======== =========
$4,748 11% $0 $0 . $4,748 11%
29 1
11% $0 (100%) $0 $4,497 ( 3%)
2 $4,497
$9,245 $0 $0 $9,245
29
== 37% $8,647 24%
30 1 $4,061 12% $0 $4,586
143
31 $7,907 $0 $0 $7,907
32 $9,245 $0 $0 $9,245
37 $4,204 $0 $0 $4,204
144
Leg Time DOC A321 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot
B737 A321 B737 A321 B737 A321 B737
Leg i lme UOC A^Zl Diff.DOC Opp.Cost Dif f.Opp Pos.Cost Diff.Pos Total Diff.Tot.
B737 A321 B737 A321 B737 A321 B737
10 $38,325
=== ======== ========= ======== ======== ========= ===:======
11 1 $5,113 24% $0 $0 . $5,113 24%
2 $4,842 25% $0 $0 $4,842 25%
3 $4,587 25% $0 $0 . $4,587 25%
4 $4,346 25% $0 $0 . $4,346 25%
5 $4,118 25% $0 $0 . $4,118 25%
6 $3,903 25% $0 (100%) $0 $3,903 12%
Leg Time DOC A321 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
B737 A321 B737 A321 B737 A321 B737
11 7 $3,306 $0 $0 $3,306
8 $3,133 $0 $0 $3,133
9 $2,970 $289 $0 $3,259
10 $2,816 $702 $0 $3,518
13 $9,173 $0 $0 $9,173
========= =========== =============== =============== ===========
14 10 $3,221 $1,053 $879 $5,153
11 $3,055 $1,609 $668 $5,332
12 $2,897 $2,107 $472 $5,476
Leg Time DOC A320 Opp.Cost A320 Pos.Cost A320 Total A320
17 $2,571 $0 $0 $2,571
18 12 $2,523 $0 $0 $2,523
18 $2,523 $0 $0 $2,523
===========
19 12 $2,523 $0 $161 $2,684
21 $17,606 $9 ,046
========= =========== =======:======== =============== ===========
22 6 $4,764 $0 $0 $4,764
7 $4,513 $0 $0 $4,513
8 $4,276 $0 $0 $4,276
9 $4,053 $433 $0 $4,486
31 $13,462 $0 $0 $13,462
========= =========== =============== =============== ===========
32 3 $4,259 $0 $0 $4,259
152
4 $4,034 $0 $0 $4,034
5 $3,822 $441 $0 $4,263
6 $3,621 $1,527 $0 $5,148
3 $3,982 $0 $0 $3,982
$3,982 $0 $0 $3,982
Leg Time DOC B737 Diff.DOC Opp.Cost Diff.Opp Pos .Cost Diff .Pos Total Diff.Tot
A320 B737 A320 B737 A320 B737 A320
13 $8,226 $0 $0 $8,226
$66,730 $0 $106056
15 $39,326
Leg Time DOC B737 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
A320 B737 A320 B737 A320 B737 A320
18 $2,251 $0 $0 $2,251
$12,924 $0 $27,048
29 $14,124
____-
$2,738 ( 29%) $5,997 ( 20%)
30 3 $3,259 ( 11%) _ $0
$0 $2,464 ( 30%) $5,549 ( 20%)
4 $3,085 ( 11%)
$0 $2,209 ( 31%) $5,130 ( 21%)
5 $2,921 ( 11%)
COST COMPONENT ANALYSIS
Absolut Values and Relative Differences to the Solution Aircraft Type
-Values in Brackets are Negative-
Leg Time DOC B737 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
A320 B737 A320 B737 A320 B737 A320
31 $12,031 $0 $0 $12,031
37 $3,571 $0 $0 $3,571
(
10%) $4,722 526% $0 $8,693 68%
38 2 $3,971 (
10%) $5,723 179% $0 $9,482 52%
3 $3,759 (
156
Leg Time DOC A321 Diff.Doc Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot
A320 A321 A320 A321 A320 A321 A320
11 $13,684 $0 $0 $13,684
12 7 $3 699 12% $6,970 ( 25%) $2, 940 117°/ $13, 609 ( 2%)
8 $3 507 12% $7,374 ( 22%) $2, 655 133°/; $13,536 ( 2%)
9 $3 325 12% $7, 722 ( 20%) $2, 389 154°/ $13, 436 ( 1%)
10 $3 153 12% $8, 020 ( 18%) $2, 140 182°/ $13, 313 ( 1%)
17 $2,879 $0 $0 $2,879
18 $2,827 $0 $0 $2,827
======== =========
19 12 $2,827 12% $0 $1,619 906% $4,446 66%
20 $2,879 $0 $0 $2,879
22 $19,690 $0 $0 $19,690
____ _____
29 3 $4,763 12% $0 $0 . $4,763 12%
4 $4,512 12% $0 $0 . $4,512 12%
5 $4,275 12% $0 (100%) $0 . $4,275 0%
6 $4,052 12% $0 (100%) $0 $4,052 ( 21%)
29 $17,602 $0 $0 $17,602
:=======
30 3 $4,074 12% $0 $5,761 50% $9,835 31%
4 $3,861 12% $0 $5,336 52% $9,197 32%
5 $3,659 12% $0 $4,937 54% $8,596 33%
COST COMPONENT ANALYSIS
Absolut Values and Relative Differences to the Solution Aircraft Type
-Values in Brackets are Negative-
Leg Time DOC A321 Diff.Doc Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
A320 A321 A320 A321 A320 A321 A320
31 $15,063 $0 $0 $15,063
32 $17,602 $0 $0 $17,602
37 $4,453 $0 $0 $4,453
:== ======== ========= ======== =========
2 $4,942 12% $0 (100%) $0 . $4,942 ( 4%)
38
3 $4,682 12% $0 (100%) $0 . $4,682 ( 25%)
160
38 $9,624 $0 $0 $9,624
161
1 $7,628 $0 $0 $7,628
2 $7,216 $0 $0 $7,216
3 $6,828 $0 $0 $6,828
4 $6,461 $0 $0 $6,461
5 $6,116 $0 $0 $6,116
6 $5,791 $0 $0 $5,791
7 $5,483 $0 $0 $5,483
8 $5,193 so $0 $5,193
9 $4,920 <
£419 $0 $5,339
10 $4,662 $1,431 $0 $6,093
11 $4,418 $2,340 $0 $6,758
12 $4,187 $3,153 $0 $7,340
Leg Time DOC A321 Opp.Cost A321 Pos.Cost A321 Total A321
4 7
$4,632 $0 $0 $4,632
8
$4,389 $0 $0 $4,389
9
$4,159 $0 $0 $4,159
10 $3,943 $0 $0 $3,943
^ $3,738 $0 $0 $3,738
12 $3,544 $0 $0 $3,544
$58,168 $0 $0 $58,168
=========== =============== =============== ===========
1 $6,425 $0 $539 $6,964
2 $6,081 $0 $150 $6,231
3 $5,756 $0 $161 $5,917
4 $5,450 $0 $161 $5,611
5 $5,162 $0 $161 $5,323
6 $4,889 $0 $161 $5,050
7 $4,632 $0 $161 $4,793
8 $4,389 $0 $161 $4,550
9 $4,159 $0 $161 $4,320
10 $3,943 $0 $161 $4,104
11 $3,738 $0 $161 $3,899
12 $3,544 $0 $161 $3,705
6 1 $5,469 $0 $0 $5,469
2 $5,179 $0 $0 $5,179
3 $4,905 $424 $0 $5,329
4 $4,647 $1,834 $0 $6,481
5 $4,402 $3,099 $0 $7,501
6 $4,171 $4,232 $0 $8,403
7 $3,954 $5,244 $0 $9,198
8 $3,748 $6,143 $0 $9,891
9 $3,553 $6,940 $0 $10,493
10 $3,369 $7,643 $0 $11,012
11 $3,195 $8,261 $0 $11,456
12 $3,030 $8,799 $0 $11,829
11 $5,828 $0 $0 $5,828
_________ =========== =============== ===========:==== ===========
12 11 $2,991 $8,271 $V ,907 $13,169
12 $2,837^ $8,480 SI.,690 $13,007
IPCJ TirnP POP &-S21 ODD fo<?t A^?1 Po* Co*t A3?1 TotM W1
21 $12,223 $5(,192
=========
22 10 $4,294 $0 $0 $4,294
11 $4,070 $0 $0 $4,070
12 $3,859 $0 $0 $3,859
22 $12,223 $0 $0 $12,223
25 1 $5,284 $0 $0 $5,284
2 $5,004 $0 $0 $5,004
3 $4,739 $371 $0 $5,110
4 $4,490 $1,605 $0 $6,095
5 $4,254 $2,712 $0 $6,966
6 $4,032 $3,704 $0 $7,736
7 $3,821 $4,589 $0 $8,410
8 $3,622 $5,376 $0 $8,998
9 $3,434 $6,074 $2,490 $11,998
10 $3,257 $6,689 $2,237 $12,183
11 $3,089 $7,229 $2,000 $12,318
12 $2,930 $7,701 $1,779 $12,410
27 $33,308 $0 $0 $33,308
$17,353 $0 $0 $17,353
=========== =============== =============== ===========
7 $3,840 $0 $0 $3,840
8 $3,640 $0 $0 $3,640
9 $3,451 $401 $0 $3,852
10 $3,273 $1,371 $0 $4,644
11 $3,104 $2,242 $0 $5,346
12 $2,944 $3,021 $0 $5,965
35 4 $4,435 $0 $0 $4,435
5 $4,203 $0 $0 $4,203
6 $3,983 $682 $0 $4,665
7 $3,775 $1,838 $0 $5,613
8 $3,579 $2,875 $0 $6,454
9 $3,393 $3,803 $0 $7,196
10 $3,218 $4,630 $0 $7,848
11 $3,052 $5,365 $0 $8,417
12 $2,895 $6,015 $0 $8,910
37 4 $4,219 $0 $0 $4,219
5 $3,999 $0 $0 $3,999
6 $3,790 $0 $0 $3,790
7 $3,592 $0 $0 $3,592
8 $3,406 $0 $0 $3,406
9 $3,229 $0 $0 $3,229
10 $3,063 $0 $0 $3,063
11 $2,905 $0 $0 $2,905
12 $2,756 $0 $0 $2,756
37 $30,959 $0 $0 $30,959
38 $4,435 $0 $0 $4,435
$4,203 $0 $0 $4,203
$3,983 $682 $0 $4,665
$3,775 $1,838 $0 $5,613
168
Leg Time DOC B737 CJiff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.f5os Total Diff.Tot.
A321 B737 A321 B737 A321 B737 A321
=== ======
3 1 $4,402 : 20%) $11,507 846% $0 $15,909 138%
2 $4,166 ; 20%) $12,343 339% $0 $16,509 107%
3 $3,943 <[ 20%) $13,070 208% $0 $17,013 86%
4 $3,733 ; 20%) $13,698 148% $0 $17,431 71%
5 $3,534 : 20%) $14,234 113% $0 $17,768 60%
6 $3,347 (: 20%) $14,688 91% $0 $18,035 52%
7 $3,170 (: 20%) $15,067 76% $0 $18,237 45%
8 $3,002 <: 20%) $15,376 64% $0 $18,378 40%
9 $2,844 (: 20%) $15,623 55% $0 $18,467 36%
10 $2,695 <> 20%) $15,814 48% $0 $18,509 32%
11 $2,554 ( 20%) $15,952 43% $0 $18,506 29%
12 $2,421 ( 20%) $16,045 38% $0 $18,466 26%
Leg Time DOC B737 C>iff .DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
A321 B737 A321 B737 A321 B737 A321
II
1
<A
15 10 $3,296 ( 19%) $9,667 87% $12,963 40%
o
11 $3,124 ( 19%) $9,902 72% $0 $13,026 35%
12 $2,961 ( 19%) $10,093 62% $0 $13,054 32%
Leg Time DOC B737 Diff.DOC Opp. Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
A321 B737 A321 B737 A321 B737 A321
27 $26,683 $0 $0 $26,683
Leg Time DOC B737 Diff.DOC Opp.Cost Diff.Opp Pos .Cost Diff.Pos Total Diff.Tot
A321 B737 A321 B737 A321 B737 A321
Leg Time DOC B737 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot
A321 B737 A321 B737 A321 B737 A321
Leg TIME DOC 320 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
A321 A320 A321 A320 A321 A320 A321
Leg TIME DOC 320 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
A321 A320 A321 A320 A321 A320 A321
Leg TIME DOC 320 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total D ff.Tot.
A321 A320 A321 A320 A321 A320 A321
$22,941 $52,714
26 $29,773 $0
COST COMPONENT ANALYSIS
Absolut Values, and Relative Differences to the Solution Aircraft
-Values in Brackets are Negative-
Leg TIME DOC 320 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
A321 A320 A321 A320 A321 A320 A321
27 $29,773 $0 $0 $29,773
== ========= ==_======
29 7 $3,432 11%) $2,502 . $0 $5,934 55%
8 $3,253 11%) $3,375 . $0 $6,628 82%
9 $3,084 11%) $4,153 936% $0 $7,237 88%
10 $2,924 11%) $4,845 253% $0 $7,769 67%
11 $2,773 11%) $5,459 143% $0 $8,232 54%
12 $2,630 11%) $5,999 99% $0 $8,629 45%
- $0 $12,366 $27,848
30 $15,482
=== ======== ========= ===== = = = = =____————
31 7 $2,936 ( 11%) $0 . $0 $2,936 ( 11%)
181
31 $15,482 $0 $0 $15,482
Leg TIME DOC 320 Diff.DOC Opp.Cost Diff.Opp Pos.Cost Diff.Pos Total Diff.Tot.
A321 A320 A321 A320 A321 A320 A321
37 $27,654 $0 $0 $27,654