aircraft design for low cost ground … design for low cost ground handling - the final results of...

AIRCRAFT DESIGN FOR LOW COST GROUND HANDLING -THE FINAL RESULTS OF THE ALOHA PROJECT

P. Krammer∗, O. Junker∗∗, D. Scholz∗∗Aero - Aircraft Design and Systems Group, Hamburg University of Applied Sciences, Hamburg,

Germany∗∗Airport Research Center GmbH, Aachen, Germany

Keywords: aircraft design, ground handling, direct operating costs

Abstract

A significant reduction in turnaround time andcosts can be achieved by changing the condi-tions of critical turnaround processes throughan adapted aircraft design. The considera-tion of foldable passenger seats together witha continuous cargo compartment yields into ashoulder wing aircraft configuration with en-gines at the tail. The resulting enhancements indis/embarking and off/loading yield, under thepremise of masses and performance identical toan Airbus A320-like baseline aircraft, a reductionof 1.3 %... 4.0 % in total DOC per seat-kilometer.However, due to the necessary adaptations of theaircraft configuration, overall aircraft masses andfuel consumption increase. The predicted bene-fits in turnaround time and costs are not able tocompensate this impact on the DOC elements,which results into an increase of total DOC perseat-kilometer.

1 Introduction and Motivation

During the past years, the predicted annualgrowth rate of 5 % in air traffic gained signif-icance due to capacity limitations and environ-mental considerations [1, 2]. One aspect to in-clude in the judgment of new aircraft concepts,that are able to serve the predicted demand, isthe aircraft-airport interface. The Airbus A380-800 is restricted in wing span and length dueto current airport terminal limitations [3]. Like-

wise, ground clearances, turning radii, turn pathsand Load Classification Numbers (LCN) mustbe taken into consideration during aircraft designand assessment. But not only taxiing, maneuver-ing and parking is crucial for an efficient airportoperation, the turnaround itself demands a num-ber of processes and coordination of many partiesinvolved: the airport, the airline and the groundhandling company. Experts in the field believethat nowadays turnarounds are already optimizedand therefore limited in a further efficiency in-crease [4]. This might be due to the fact that theaircraft remained as it is and only the ground han-dling processes together with used ground han-dling equipment have been adapted to serve fora more efficient turnaround. In this context, thequestion arises, if there would be a further poten-tial in efficiency increase due to adaptations ofthe aircraft design itself which is task of the jointresearch project Aircraft Design for Low CostGround Handling (ALOHA) among the follow-ing partners:

• Hamburg University of Applied Sciences(HAW Hamburg) - acting as project leader

• Airbus Operations GmbH (Future ProjectOffice)

• Airport Research Center GmbH (ARC)

• Hamburg Airport GmbH (Ground Han-dling Division)

ALOHA aims at identifying technologies and air-craft design adaptations that are exhibiting the

1

P. KRAMMER∗, O. JUNKER∗∗, D. SCHOLZ∗

potential of an efficient ground handling in termsof costs and time demand. The HAW Ham-burg involvement in the ALOHA project is finan-cially supported with a grant of the FHprofUndprogramme from the Federal Ministry of Ed-ucation and Research. The grant is adminis-tered by the Arbeitsgemeinschaft industriellerForschungsvereinigungen Otto von Guericke e.V.(AiF). The duration of ALOHA is 3 years and 2months. It started in November 2007.

2 The Airlines’ Perspective

With the rise of fuel costs and fees that are addi-tionally nowadays related to environmental con-siderations, airlines strive to reduce costs in ar-eas and segments that are, in contrast to aforementioned ones, being able to get influenced.What remains as an airline authority can be ascer-tained by summing up the entire operating costsof an aircraft as part of the Direct Operation Costs(DOC).

CDOC =CDEP +CINT +CINS +CF+

CM +CC +CFEE(1)

CFEE =CFEE,LD +CFEE,NAV +CFEE,GND (2)

The costs C incurred due to depreciationCDEP and interest CINT are primarily related tothe purchase price of the aircraft. There is also arestricted influence on insurance costs CINS cov-ering hull damage and hull loss. As has beennoticed over the past years, fuel costs CF can-not be influenced at all. Maintenance costs CMcan be lowered by bringing into a play a fleetof homogeneous aircraft types to reduce costsof spare parts (e.g. through quantity discounts)and/or aircraft type trainings. Crew costs CCmust be kept equal to standardized levels be-cause otherwise airlines run the risk of crewstrikes [5]. Landing fees CFEE,LD as well as nav-igation fees CFEE,NAV are depended on aircraftmass, region and airport and are thus subject tothe offered destinations. Ground handling feesCFEE,GND are (predominantly) dependent on theactions required during a turnaround and negoti-ated with local ground handling companies or air-ports through Service Level Agreements (SLA).

As a consequence, to remain competitive, airlineshave nowadays only limited possibilities to re-duce their operating costs, which are, by reducing(1) and (2):

∆CDOC ≈δCM

δx1·∆x1 +

δCFEE,GND

δx2·∆x2 (3)

Although limited, the potential in reducing costshas been successfully exploited by well estab-lished Low Cost Airlines (LCA) such as South-west Airlines and Ryanair. It is therefore likelythat the low cost airlines segment will continueto grow, claiming an increasing market share oftravel by air. Their success has sprouted a globalinterest of all airlines and puts them under pres-sure to catch up and remain competitive. One ofthe key enablers of LCA are adapted turnarounds[6, p. 22] as can be clearly seen in (3), par-tially only in combination with secondary air-ports though [7].

3 Problem Stating

An improvement in ground handling can lead totwo principal effects: a reduction in ground han-dling costs, and an increase in aircraft utilization.

3.1 Ground Handling Cost Reduction

In principal, a cost reduction at one single groundhandling process can be noticed on the totalground handling costs and thus, on the total air-line related costs. One of the overall issues is thehigh dependency on other processes, resourcesand/or stakeholders. Thus, by reducing the in-terfaces between the aircraft and the airport ter-minal, a reduction in required Ground SupportEquipment (GSE) and ground handling staff canbe accomplished that would further reduce asso-ciated costs as well as e.g. the potential of delays.This means that the aircraft has to become moreautonomous (i.e. getting independent of exter-nal GSE) such as including an autonomous pushback system and on-board air stairs. On the con-trary, such equipment leads to an increase in air-craft weight. Likewise, the aircraft must be de-signed to accommodate for the new technology.

2

AIRCRAFT DESIGN FOR LOW COST GROUND HANDLING - THE FINAL RESULTS OF THEALOHA PROJECT

Therefore, as improvements to ground handlingoperation always aim at reducing turnaround timeand ground handling costs, it needs a close lookto find out if improvements to ground handlingoperations also reduce the total DOC. This hasto be done because in some cases, a reductionin ground handling costs increases the aircraftweight and delivery price, which leads to draw-backs in cruise performance and other DOC costitems such as fuel costs and depreciation (1).

3.2 Ground Handling Time Reduction

In contrast to plausible cost reductions, the re-duction in time of one single ground handlingprocess might not essentially lead to an over-all reduction in turnaround time. This is be-cause only the ground handling processes thatare on the critical path are influencing the over-all turnaround time. Processes not on the criticalpath are running simultaneously to critical pathprocesses but do not dependent on predecessors.Secondly, an overall reduction in turnaround isnot directly linked to an increase in utilization.This effect might only be noticed if the reductionin turnaround time throughout the day may leadto a possibility of a further flight during the con-sidered daily availability. (4) emphasizes this byshowing the parameters involved:

Ud, f = t f ·Ad

t f + ta + tt= t f ·n f ,d (4)

whereUd, f daily utilization [h/d]Ad daily availability [h]t f flight time [h]ta turnaround time [h]tt taxi time [h]n f ,d number of flights per day [-] integer

The utilization U of and aircraft is a parame-ter with a strong influence on DOC and indicatesthe efficiency of an airline’s operation. Airlinesconsistently strive after maximizing their utiliza-tion in order to distribute their fixed costs (de-preciation, interest, insurance) over an increasingnumber of flights, which is equivalent to a rela-tive cost reduction of each individual flight. To

do so, the parameters involved in (4) must be sur-veyed individually: The flight time t f is depen-dent on flight plan and therefore not a parameterto be adapted. Likewise the taxi time tt whichis mainly dependent on the airport layout. Thedaily availability Ad of an aircraft depends onmany external things: flight plan, airport nighttime restriction, etc. The only parameter remain-ing for increasing the utilization is the turnaroundtime ta. Hence the question arises: How much inturnaround time reduction is needed to increasethe daily utilization Ud, f and thus the annual uti-lization Ua, f of an aircraft? Since the number offlights per day n f ,d is an integer number, (4) mustbe rewritten to obtain the relative daily utilizationUd, f ,rel in [%]:

Ud, f ,rel =t f · bn f ,dc

Ad=

⌊Ad

t f + ta + tt

⌋· t f

Ad(5)

bn f ,dc= max{

k ∈ Z|k ≤ n f ,d}

(6)

Thus, the relative daily utilization increasesonly if the number of flights increases by oneflight. This yields into the following requirement:

Ad

t f + ta + tt= bn f ,dc+1 (7)

Figure 2 depicts the relative daily utilizationover the flight time for two different standardturnaround times by applying (5). The relativedaily utilization continuously increases with anincreasing flight time up to a point where thenumber of flights suddenly decreases by one dueto the fact of another flight in between. Thischaracteristic is more significant at short rangeflights. Figure 3 depicts the necessary reduc-tion in turnaround time to increase the utiliza-tion for different standard turnaround times byapplying (5) and (6). Accordingly, the higherthe flight time the higher the required turnaroundtime reduction for increasing the utilization (notethat a reduction of turnaround time above 30min is physically not possible since the standardturnaround time was set at 30 min or 20 min re-spectively). With shorter turnarounds through-out the day, the absolute impact on the rela-tive utilization is slightly lower but achieved at

3


a lesser (absolute) number of minutes in requiredturnaround time reduction.

A reduction in turnaround time can be sig-nificant and depends primarily on the ratio offlight time to turnaround time. The possibil-ity to achieve a higher utilization is higher forshort range flights. With Ad = 18 h, t f = 1 h,ta = 30 min, and tt = 10 min, a reduction of about1.8 min in turnaround time ta must be achieved toincrease the utilization. This value increases to5.7 min for flights with 2 hours flight time.

4 Analysis of the Current Turnaround Situ-ation

To get a broader perspective of the various issuesinvolved in the daily ground handling of an air-craft, interviews have been conducted with ex-perts in the field. To later judge individual air-craft design proposals in a correct way, data ofreal turnarounds has been collected and analyzed.

4.1 Expert Interviews

In order to address the real issues that are facedin the daily aircraft ground handling, experts inthe field of ground handling procedures, airlinebusiness strategies and ground handling equip-ment have been interviewed. The informationcollected (out of a total of ten interviews con-ducted) has been transferred into significant andshort statements and separated for each groundhandling process [4].

Results emphasize the need of a betterdoor positioning with respect to ground han-dling equipment dimensions through lowered sillheights and door clearances (e.g. cargo loadingand catering processes are often restricted by theengine nacelles. In addition to this, the engine in-let is highly susceptible in the event of secondarydamages. The wing root is e.g. interfering withthe passenger boarding bridge at the second doorof the Airbus A321).

Furthermore, Center of Gravity (CG) limita-tions decrease the flexibility of cargo loading andenforce many processes that are not optimal: e.g.passing of loose baggage along the cargo hold

into the rear end of the cargo compartment; useof cargo nets requires tying of cargo nets which istime consuming and the position of it is decisivefor the CG location or cargo loading in a certainsequence (unloading from back to front and load-ing from front to back).

The process of refueling from the fuel truckramp is rather often accomplished than in theconventional way with ladder and rolled out fuelhose which needs more time for preparation.Reasons for that are either that the fuel truck can-not park directly under the aircraft wing becauseof the low wing height (Boeing B737) or, as itis often the case, the fuel truck parking near thecargo holds is interfering with the loading and offloading process.

Experts in the field additionally pointed outthat there is a wastage of cargo volume by makinguse innovative ground handling equipment suchas sliding carpets. The pushback must be con-sidered as a critical process since it can lead todelays and the missing of slots. A lower accessi-bility can be noticed for underwing deicing withwings located close to the ground so that is usu-ally accomplished by hand. Bigger GSE is usedfor ground handling than it is actual necessary forthe type of aircraft at the airport to serve moretype of aircraft with the same purchased GSE.Using larger GSE than necessary is often not op-timal. Obviously, the capacity and flexibility ofground handling activities is limited by the num-ber of available aircraft doors.

4.2 Turnaround Process Analyses

168 turnarounds of typical single-aisle aircraftin the short to medium range segment at fourdifferent airports have been video-tape recordedand analyzed. Collected data has been pre-pared to undergo mathematical regressions (e.g.number of passengers versus disembarking time)and statistical evaluations of all possible sets ofturnaround parameters. Where obvious, outliershave been deleted to get a representative datasample. The evaluation of ground handling pro-cesses is thus based on measures of central ten-dency and dispersion as well as probability den-

4


sity distributions.Results showed that only a few processes cor-

relate linearly. A higher order mathematical re-gression would not deliver better results sincecollected data is much dispersed. This is due tothe fact that ground handling processes necessi-tate much of activities and are therefore involvingand depending on various parameters. Althougha total of 96 parameters have been recorded andare thus available in the data sample, a certainnumber of influences that might even be hardto capture, remain unconsidered. For instance,Figure 1 (top left and right) depicts the numberof passengers over disembarking and embarkingtime. As can be seen, the disembarking time cor-relates almost linearly with the number of pas-sengers. In contrast to that, the embarking time ismuch dispersed, prohibiting a successful regres-sion analysis.

Furthermore, in some cases the recorded pro-cess times cover more activities than the actualconsidered one. For instance, as the refuelingprocess has been captured by a time stamp of fuelhose connecting and disconnecting, the definitefuel flow starts a few moments later. In this case,the data sample is representative to calculate totalrefueling time rather than fuel flow rates.

According to statistical evaluations, most ofthe processes are exhibiting a log-normal distri-bution characteristic (9) such as the disembark-ing process as shown in Figure 1 (middle left).This is because the probability of a value bel-low a certain (minimum) time is close to zero dueto the fact that the process cannot be carried outin such a relatively short time. The slope of theprobability function then rises up very quickly upthe mode where it then decreases with a nega-tive but lesser slope than before. This means thatthe probability of a disembarking time below themode is very low in contrast to the probability ofa disembarking time above the mode. Thus, themode in this example could be considered as anabsolute minimum for later comparisons. For theprocess of embarking however, the Normal dis-tribution (8) is in better agreement with the datasample (Figure 1 middle right) as the probabil-ity distribution shows a symmetrical characteris-

tic about the mode of around 9 min in embarkingtime. As a result, the probability of achieving anembarking time below or above the mode is sim-ilar and emphasizes the hypothesis that the em-barking process is dependent on various parame-ters and cannot be captured solely by the numberof passengers.

f (x) =1

σ√

2Πexp

(−1

2

(x−µ

σ

)2)

(8)

f (x) =

1

xσ√

2Πexp(−(Ln(x)−µ)2

2σ2

), x > 0

0, x≤ 0(9)

To evaluate the influence of other (recorded) pa-rameters on e.g. disembarking time, the Pearsonproduct-moment correlation coefficient (PMCC)has been calculated according to [8]:

r =∑(xi− x)(yi− y)√

∑(xi− x)2 ·∑(yi− y)2(10)

with the measured value of xi of the character-istic X1 and yi of the characteristic X2 at the i-th individual. PMCC values help to identify lin-ear correlations between couples of selected datasamples rather than expressing any curve pro-gression other than linear. Figure 1 (bottom) de-picts a number of PMCC values of different (pre-selected) parameters related to the dis/embarkingprocess. What can be noticed is that disembark-ing time correlates almost linearly (r > 0.5) withthe aircraft model (A319, A320, A321, B737,B738), the number of aircraft seats, and, as pre-viously mentioned and depicted in Figure 1 (topleft), on the number of disembarking passengers.For the embarking process however, no represen-tative linear correlation could be found. Inter-esting to see however is that the embarking pro-cess does not depend on aisle width, seat pitch orstage length. Some pairs show a very good linearcorrelation e.g. seat pitch versus airline businessmodel where it has to be kept in mind that othersare a reason of logic interrelations such as num-ber of aircraft seats versus number of passengers,aircraft type or seat pitch.

5


5 Aircraft Design Synthesis and Analyses

As can be seen in Figure 2 aircraft that aremost affected by turnaround time are short andmedium range aircraft, in particular the BoeingB737 and the Airbus A320. The B737 was devel-oped in the 1960’s, the A320 in the 1980’s. Thisexplains why requirements of nowadays groundhandling operations (comparable to that of lowcost airlines) were not considered in the designof the B737 and A320. The manufacturers havealready announced successors of the B737 andA320. It is therefore of great interest to comeup with adapted aircraft design proposals that areadditionally featuring enhanced ground handlingcapabilities.

With this in mind, the twin-engined AirbusA320 has been chosen as the baseline aircraft forlater comparisons and analyses with the proposedaircraft designs.

5.1 Aircraft Design Process in PrADO

Basis of aircraft design analysis and multidis-ciplinary design optimizations is the Prelimi-nary Aircraft Design and Optimisation programPrADO [9, 10], developed by the Institute ofAircraft Design and Lightweight Structures ofthe Technische Universität Braunschweig. Thecore of the program reflects a set of design mod-ules, representing each relevant discipline in-volved in the preliminary aircraft design pro-cess that are run until overall aircraft massesare converged. The program can easily be ex-tended by adding new modules to take accountfor further aspects that are gaining significance[11]. Likewise, modules not necessary for a de-sign analyses can be deactivated to lower com-putational effort. All modules are communicat-ing with each other only through a data manage-ment system which accesses thematically sorteddatabases. The databases are initialized throughdata of the transport mission, a basic paramet-ric description of the configuration layout, con-straints and design targets. Additionally, differ-ent methods for a specific task are available re-flecting user-defined requirements such as level

of detail, method of analysis, etc.In context of the ALOHA project, the mod-

ule DOC becomes significant. The PrADOmethod selected for DOC computation is theDOC method according to Association of Euro-pean Airlines (AEA) [12] due to its completenessand public availability. The method of AEA wasestablished in 1989. For later comparison andanalysis with available cost indices, a compensa-tion for inflation must be provided to account forthe predominantly financial condition of all theyears in between:

kINF =nyear

∏nmethod

(1+ pINF) (11)

According to the U.S. Department of Labor kINFamounts to 1.79 for the period of 1989 to 2010with average consumer price indexes as a calcu-lation basis [18].

The design method in PrADO has been vali-dated (Table 1) through a re-design of the base-line aircraft (Figure 6). Overall aircraft massesare in good agreement with data out of [14].Take-off field and landing-field length accord-ing to FAR requirements as well as the approachspeed are subject to flight mission simulations offlanding and take-off and predicted with a devia-tion of about 13 %. However, on this basis, therequirements of further adapted aircraft designshave been set to values of Table 2, compensat-ing the deviations subject to flight mission simu-lations.

5.2 Aircraft Design Proposals and Analysis

The aircraft design under investigation is anAirbus A320 aircraft adapted for ground han-dling operations. The analysis of the cur-rent turnaround situation (Chapter 4) showed,that a shoulder wing aircraft could exhibitthe capability of time-reduced and cost-reducedturnarounds. The configuration to be investigatedis therefore an Airbus A320-like configurationwith a shoulder wing instead of a conventionallow wing.

In the case of a shoulder wing, the landinggear integration becomes challenging. On the

6


contrary, the wing can be designed without a kinkwhich yields into a simple (and single) taperedwing design. To lower the length of the land-ing gear, the fuselage should be kept close to theground which additionally eases ground handlingoperations due to lowered sill heights. With thisin mind the tail clearance for appropriate take-off rotation becomes decisive. A remedy can befound be additionally locating the engines at thetail of the fuselage, shifting the aircraft CG andthus the wing position and landing gear adjust-ment (relative to the fuselage) backwards. Thefree area obtained below the aircraft wings addi-tionally eases ground handling processes such asrefueling or loading and offloading because therewill be no interferences with the engine nacellesanymore. The calculation of the wing positiontowards an optimum overall CG has been imple-mented in PrADO, by solving an equilibrium ofmomentum around the leading edge of the MeanAerodynamic Chord (MAC) of the wing group(WG) and the fuselage group (FG):

x = xFG−∆xCG +mWG

mFG(∆xCG,WG−∆xCG)

(12)withx longitudinal axisx x coordinate of leading edge at MAC∆xCG,.. distance between CG and x∆xCG e.g. set to 0.25 · cMACm.. mass of wing or fuselage group

Although optimized in wing positioning, thestretched fuselage of the shoulder wing config-uration will get heavier. However, the questioncan be raised, if the shoulder wing configurationis able of compensating its structural and aero-dynamic penalties through a simple tapered wingdesign. Figure 4 summarizes and visualizes theresults of an calculated Pareto-optimal boundarywith design variables x of aspect ratio (AR), wingreference area (Sre f ) and taper ratio (λ):

minx∈ℜn

=

{∣∣∣∣fmW (x)fmF (x)

∣∣∣∣g(x)≤ 0}[13] (13)

with the objective functions of minimum wingmass fmW (x) and minimum fuel mass fmF (x) (cal-culated for the selected reference mission). The

overall tendency of the selected objective func-tions is depicted according to the theory: the bet-ter the aerodynamic quality – the higher the as-pect ratio or wing span – the higher the wingbending moment and thus the structural weightof the wing. In contrast to that is the wing -very short in span - that is very light in weightbut exhibits a poor aerodynamic quality. Whatcan be seen is that the fuel mass is not decreasinganymore after a certain point (around the wingmass of 13 t) which means that any further in-crease in wing span is not favorable. This is dueto the structural penalty of the wing which is -after this point - higher than the aerodynamic im-provement.

Figure 4 shows additionally where the wingof the baseline aircraft would be located in termsof wing and fuel mass for the same flight mis-sion. The point is situated below the Pareto-optimal boundary due to the fact of lower oper-ating empty mass which mainly results out of alower fuselage mass (11.7 t versus 9.3 t of fuse-lage mass). This leads to the conclusion that theshoulder wing configuration is not able of com-pensating its structural and aerodynamic penal-ties through a simple tapered and optimized wingdesign.

When it comes to optimizations with DOC asa target function, the wing design point is mov-ing along the Pareto-optimal boundary accordingto the cost parameters selected. To come up witha final wing design, the shoulder wing configura-tion has been optimized towards minimum DOCper seat and per kilometer with cost parametersselected and listed in Figure 8. The resultingpoint is slightly above the Pareto-optimal bound-ary (Figure 4). This is because of the fact thatthe fuselage mass dependency on wing mass hasbeen neglected during calculations. What can ad-ditionally be noticed is that the wing has beenoptimized towards a more aerodynamic efficientdesign compared to the baseline aircraft. This isdue to the fact that the DOC parameters selectedfor optimization do incorporate higher fuel priceswhich was obviously not the case when the Air-bus A320 was designed back in the 1980’s.

The proposed aircraft designs additionally

7


feature foldable passenger seats [15] next to theaisle for a faster dis/embarking. The increasein fuselage mass can therefore additionally beblamed on the heavier foldable passenger seats(40.0 kg instead of 29.9 kg of a triple economyseat bench).

The idea of reducing the time of processes onthe critical path of the turnaround, has been fur-ther extended by designing a six abreast, twin-aisle shoulder wing aircraft (Figure 6). The em-pennage of the shoulder wing configuration couldbe adapted towards an H-tail to allow for en-gine noise shielding effects (compare [16]). Thisis however not a primary task of the ALOHAproject.

6 Ground Handling Analyses of ProposedAircraft Designs

In order to assess the aircraft design proposalsin terms of ground handling operations, groundhandling scenarios have been developed that havebeen reproduced in the simulation environmentCAST Ground Handling. The process orientedcost calculation has been chosen to accuratelypredict the difference in ground handling costs.

6.1 Ground Handling Scenarios

All collected data from turnaround analyses(Chapter 4.2) has been fed into turnaround Ganttcharts that are thus based on realistic turnaroundparameters. It has been noticed, that turnaroundprocesses are primarily dependent on the air-line business model (conventional versus lowcost) and the parking position (terminal or re-mote apron). In order not to loose this infor-mation through a rough averaging of the real-time turnaround parameters, four ground han-dling scenarios have been defined as depicted inTable 4. Scenarios at the terminal (I and III) re-flect a full service turnaround including a clean-ing of the cabin, potable water refilling and wastewater service as it would occur at every third orfourth turnaround on short-haul flights. A furtherdifferentiation is made between conventional andlow cost airlines (I and II vs. III and IV): use

of containers (transfer baggage) vs. bulk cargo,full catering vs. limited catering, towbarless pushback vs. towbar pusback (less GSE costs), etc.Also, a different passenger load factor betweenlow cost and conventional airlines is taken intoconsideration.

An example of the derived Gantt chart is de-picted in Figure 5. The length of each bar hasbeen scaled according to the results of the regres-sion analysis of each turnaround process wherethe lines indicate the process time with respectto the standard deviation of each turnaround pro-cess. All scenarios and derived Gantt charts arethus based on realistic turnaround data and rep-resent the basis of evaluation of adapted aircraftdesigns. Furthermore, all derived Gantt charts donot take into consideration a refueling parallel todis/embarking as this is not the case in the 168turnarounds video-tape recorded.

6.2 CAST Ground Handling

Results from spreadsheet (Gantt chart) analy-sis have a disadvantage: they will not be ableto cover requirements resulting from geometrybased component interaction. Simulations al-low to bring the different complex views togetherin one model by the use of layered modeling,each of these layers reflecting thus simulation andanalysis of the individual reality. In addition tothe layered structure of the simulation engine, itwas necessary to remain flexible during the de-velopment as well as in the modeling process.Using simulations off-the-shelf, the user will usu-ally either not be able to realize specific needscoming from the analysis or the simulation will –because it only provides general building blocks– need quite much user effort to generate sophis-ticated yet understandable results.

By using the Comprehensive Airport Simu-lation Tool (CAST), an in-house development ofthe Airport Research Center [17], as a basis fora new ground handling module, software classescan be quickly and flexibly adapted while, at thesame time, being powerful enough in fulfillingthe needs raising from discussed requirements.As a result, the ground handling part of the sim-

8


ulation engine CAST Ground Handling has beendeveloped within ALOHA and allows for simula-tion of different service arrangements of differentaircraft models.

From a user’s point of view, different aspectsof the simulation itself and the simulation resultscan be realized by quite detailed modeling abili-ties and powerful object filters which only reportrelevant model and analysis details. By quick ob-ject instantiation, the user can handle the basecomponents such as aircraft models, vehicles andlane segments to quickly reflect the turnaroundscenario that shall be analyzed. Because everysingle object can be configured individually in-cluding the definition of properties, methods andprocesses, the simulation scenario can be devel-oped and tested step-by-step, providing interimsimulation results and thus evaluation and vali-dation in every stage of the modeling, increasingthe reliability and thus the efficiency.

From the developer’s point of view, CASTprovides an object oriented simulation core,which allows a capsulated and safe development.In addition, CAST contains a sophisticated state-of-the-art 3D rendering engine including abilitiesto model textured, detailed and precise geometrydetails. As an example, it was possible to imple-ment an importing component to provide the de-tailed aircraft geometry out of PrADO, acting asan autonomous agent in the CAST engine (Fig-ure 7). This conversion of the 3D PrADO geome-try into the 3D CAST Ground Handling environ-ment allows for ground handling simulation ofdifferent aircraft designs that have been designedand pre-evaluated with PrADO. Finally, the timeand process based simulation run itself allowsto define complex object interactions in advance,that is, during the modeling process itself, theuser can describe what shall happen when theparticipating object is led into a certain state dur-ing the simulation run.

During the modeling process, CAST allowsthe user to instantiate objects of the developedclasses in a simulation model, which aggregatedreflects the desired scenario. Once connected andconfigured, the simulation runs will modify theobjects’ properties based on the natural classes’

behavior as well as the user’s configuration. Thisincludes realistic behavior of the participating ob-jects – for instance, the vehicles will move alongthe defined lanes, reflecting tow curve behaviorand loader modification (platform lifting, cargotransport, fuselage docking at predefined connec-tion points, etc.).

Once the different simulation scenarios havebeen performed, CAST allows to extract the sim-ulation results as well as process times and theirdetailed cost structure efficiently by the use ofspreadsheet software.

6.3 Process Cost Calculation in GroundHandling

Process cost calculation uses an approach that al-lows for a better assignment and control of in-direct divisions or e.g. service provider costs.This means that the costs can be imposed to theiractual product or service. Relating to each sin-gle company processes, a first orientation can befound at the value chain. The company cost cen-ters are splitting general tasks into process ori-ented activities. Process cost rates will then becalculated by assigning costs (that are subject tothe so called cost drivers) to the process activ-ities. The process oriented overheads (burdencosts) are then imposed on the products and ser-vices with the help of the derived process costrates [19, Chapter 5].

The process cost calculation (which is a fullcost pricing method) takes only into account pro-cesses with a repetitive character. This is be-cause the cost drivers can only be imposed onsuch processes. In contrast, processes with anon-repetitive character cannot be evaluated an-alytically. Processes with repetitive character arefurther subdivided into activity quantity inducedand activity quantity neutral processes. The latterones are in general calculated by being drawn upto a budget. Process cost rates (of processes witha repetitive character that are activity quantity in-duced) are calculated by means of the cost driverpCD i.e. per definition: process costs divided byprocess quantity. For processes with a repetitivecharacter that are activity quantity neutral, a fixed

9


amount allocation pconst has to be evaluated i.e.the ratio of costs of activity quantity induced andactivity quantity neutral processes which is ad-ditionally multiplied by the cost driver pCD [19,Chapter 5]. The total cost driver p is then the sumof both (14).

p = pCD + pconst (14)

CFEE,GND =np

∑i=1

nA

∑j=1

[ti, j(xops,k

)· pi, j,CD ·ni, j,RES + pi, j,const

](15)

wherenp no. of individual ground handling pro-

cesses involvednA no. of activities of the individual

ground handling process iti process time of an individual ground

handling activity j within the groundhandling process i that is, if applica-ble, a function of xops and k

xops operational parameter such as no. ofseats, no. of baggage, volume of fuelrefueled, etc.

k average rate of the operational param-eter xops

pi, j,CD cost driver of activity j within theground handling process i

ni, j,RES number of resources necessary for ac-tivity j within the ground handlingprocess i

pi, j,const fixed amount allocation for activity jwithin the ground handling process i

The process cost calculation of ground handlinghas been successfully applied and demonstratedby [20]. In principal, on every process that takespart in the ground handling of aircraft, actualcosts can be imposed. This means that the groundhandling fee is the sum of all process and ac-tivity related costs involved in the ground han-dling. Thus, (15) can be stated in general forestimating the ground handling costs. The pa-rameters xops and k can be found for the mainground handling processes out of the regressionand statistical analysis of collected turnaround

data (compare Chapter 4.2). For the cost drivers,a further data analysis becomes necessary. Somedata can be found in [20], other might be possi-ble to gather from webpage’s of ground supportequipment manufacturers, ground handling com-panies, and airports.

7 Aircraft Design Assessment and Discus-sion

At the time of writing, the analyses of the pro-posed aircraft design with the help of CASTGround Handling have not been finished (Fig-ure 7). The method of analysis is therefore re-stricted to Gantt chart analysis and process ori-ented cost calculation according to (15).

The reduction in turnaround time hasbeen predicted by adapting the processes ofdis/embarking and off/loading of derived Ganttcharts of each individual ground handling sce-nario. According to preliminary dis/embarkingsimulation results, the foldable passenger seatyields a reduction in disembarking time of 44.9 %and in embarking time of 16.6 %, both consid-ered for a one door operation (scenario I and IIIdue to the use of a passenger boarding bridge).The difference in embarking and disembarking isin agreement with the results obtained from thereal data turnaround analysis (Figure 1). It is es-timated that for a two door operation (scenarioII and IV) the same percentages in reduction ap-ply (further simulations pending). Furthermore,the cleaning time has been reduced by 1/3 dueto a better accessibility. Due to simultaneous of-floading and loading of cargo, the GSE does nothave to be changed from the aft to the forwardcargo door and vice versa. For low cost airlines,the loose baggage has to be loaded into contain-ers, before the container is loaded into the air-craft. This is considered as more time efficientand easier in operation (alternatively, a container-like box inside the cargo hold could be loadedwith loose baggage either by hand due to thereduced cargo sill height or with the help of abelt loader). After loading, the container canthan be slid to the correct CG position. As aconsequence, no weight and balance calculations

10


have to be accomplished (optimal CG positioncould be located by weight sensors at landinggear struts) and no proportioning of baggage andfreight into forward and aft cargo holds is disar-ranging the ground handling process. For all thisreasons, it is believed that the off/loading pro-cesses are not decisive for the overall turnaroundtime anymore and in non of the cases part ofthe critical turnaround path. The reductions inturnaround time are depicted in Figure 8. Thehigh percentage in turnaround time reduction inscenario II is because of the switching of the crit-ical path processes. Obviously, a further reduc-tion in turnaround time could be possible for sce-narios II and IV by making use of a three-dooroperation (not considered to the lack of empiricaldata).

The reduction in turnaround costs has beenpredicted with the help of cost parameters tobe found in [20]. Due to a turnaround time-depended amount of the total turnaround relatedcosts (≈ 70 %) an overall reduction in costs isdepicted in all scenarios (Figure 8). ScenariosI and III require crew and passenger transportsfrom the remote apron to the terminal which risesthe absolute value. For all scenarios a containerloader instead of a belt loader is needed. Thisgives two container loaders for scenarios I and IIand only one for scenarios III and IV. Due to theeasier operation, only two instead of three aircraftloaders of the ground staff are necessary. For sce-narios III and IV, the ground staff remains as it isdue to off/loading of loose baggage into the con-tainer boxes. Although the container loaders aremore expensive than the belt loaders, the over-all reduction in turnaround costs is depicted dueto the reduction in the number of ground han-dling personnel needed and their active time (e.g.cleaning).

With predicted reductions in turnaround timeand costs, DOC calculations have been per-formed for different DOC flight missions (DOCrange = 300 nm, 500 nm, and 700 nm). Withan assumed daily availability of about Ad = 16 h,and annual availability of Aa = 3750 h accordingto [12], and under the premise of overall aircraftmasses and performance identical to the baseline

aircraft, a total reduction of 1.3 %... 4.0 % in to-tal DOC per seat-kilometer can be noticed. How-ever, by implementing the concepts of foldablepassenger seats and simultaneous loading and of-floading into the aircraft design, the overall ben-efit in DOC gets lost. This is due to the higheroverall aircraft masses of the proposed aircraftdesign which rises almost all DOC elements in(1) and (2), but primarily as a result of a higherfuel consumption of about 9.5 %. A higher pur-chase price of the aircraft is also taken into con-sideration which is however not crucial. The rel-ative annual utilization yields into discrete val-ues according to (5) and (6) and is also subject toperformance calculations and therefore the flighttime. It can be seen that although the relative an-nual utilization increases by up to 19.7 %, theDOC per seat-kilometer value increases. Thus,the advantage of a better turnaround disappearsdue to the mass penalties of the proposed aircraftdesign.

Although the presented method of turnaroundanalyses takes a number of parameters into con-sideration, not all improvements could have beenassessed with respect their financial impact. Anexample of that are possibilities of delays due toturnaround issues which could lead into the miss-ing of departure slots. Such situations are becom-ing increasingly the case at highly optimized andincreased capacity airports. It is believed that bysimplifying operational turnaround processes thepossibility of delays subject to ground handlingactivities is reduced, putting the proposed shoul-der wing aircraft into another spotlight. Further-more, the potential of optimizing e.g. the fuse-lage has not been exhausted. Weight reductionswould therefore be possible.

Due to the high influence of overall aircraftmass on the total DOC, the concept of the twin-aisle shoulder wing aircraft as depicted in Fig-ure 6 has not been followed. A second prob-lem that might occur with this concept is the bot-tleneck of the passenger boarding bridge whatcould nullify the benefit of the second aisle dur-ing dis/embarking for scenarios I and II. Further-more, as the turnaround time could be reduced al-ready by a remarkable extend, a further reduction

11


would be problematic in other areas than groundhandling. As an example, the pilots itself de-mand some time to work off all required preflightchecks.

8 Summary

A short turnaround time of an aircraft is nowa-days essential for an efficient and economic op-eration of an aircraft. Although nowadays groundhandling processes are already optimized to acertain extend, new possibilities evolve by adapt-ing the aircraft design itself, providing new con-ditions for turnaround processes. In principal,an improvement in ground handling can lead toan increase in aircraft utilization and/or a re-duction in ground handling costs. The increasein aircraft utilization can only be noticed if thetotal time reduction of all turnarounds of theconsidered day enables the possibility of con-ducting another flight. A significant reductionin turnaround time can only be accomplishedby adapting the processes of dis/embarking andoff/loading. This is because the adaption of onlyone process could shift the other process onto thecritical path of the turnaround. A significant re-duction in dis/embarking time can be achieved byincorporating foldable passenger seats next to theaisle, which reduces the aisle interference. Foroff/loading of cargo and baggage, a continuouscargo compartment is suggested that enables a si-multaneous offloading or loading as well as anoptimum aircraft CG positioning by an optimumcontainer positioning in the cargo hold. Incor-porating these two enhancements in an aircraftyields into a shoulder wing aircraft configurationwith the engines located at the tail of the air-craft. Under the premise of masses and perfor-mance identical to an Airbus A320-like baselineaircraft a total reduction of 1.3 %... 4.0 % in totalDOC per seat-kilometer can be noticed. The nec-essary adaptations of the aircraft yield howeverinto an increase of overall aircraft masses due toan increase in fuselage length and the enginesat the tail. The resulting higher fuel consump-tion rises the fuel costs which are in relation tothe increase in aircraft utilization and decrease in

ground handling costs of a greater extend. Forthese reasons, the proposed shoulder wing air-craft design featuring a time and cost reducedturnaround results in a higher total DOC per seat-kilometer in comparison to the unchanged base-line aircraft with poor turnaround time capabili-ties. However, the proposed shoulder-wing air-craft configuration could reduce the possibilityof delays caused by ground handling operationswhich has not been taken into consideration inthe analysis presented.

Acknowledgments

The authors acknowledge the financial supportof the Federal Ministry of Education and Re-search (BMBF) which made this work possible.The authors gratefully acknowledge the coopera-tion of all experts to conduct interviews and theircontributions as well as the contributions fromall airlines for empirical data acquisition. Thefirst author would like to thank Axel Denglerand Sebastian Pahner from Airbus for interest-ing discussions and the exchange of ideas as wellas Oliver Crönertz for his contributions towardsground handling cost analyses.

References

[1] Advisory Council for Aeronautics Research inEurope (ACARE). European Aeronautics: A Vi-sion for 2020. ACARE, Brussels, 2001.

[2] Penner J E, Lister D H, Griggs D J, et. al.Aviation and the Global Atmosphere - Sum-mary for Policy Makers. IPCC InternationalPanel on Climate Change, Geneva, 1999. URL:http://www.ipcc.ch/ipccreports/

[3] Collins T D. Analysis of A380 Transport. De-partment of Aerospace and Ocean Engineering,Virginia Polytechnic Institute and State Univer-sity, 2010. URL: http://www.aoe.vt.edu/

[4] Krammer P, Binnebesel J, Scholz D. Anal-ysis of Interviews Performed with GroundHandling Experts. Aero - Aircraft Designand Systems Group, Hamburg Universityof Applied Sciences 2010. Technical Re-port (ALOHA_FinalReport_Expert-Interviews_2010-03-31)

12


[5] Sheahan M, Jones R. Lufthansa strike sus-pended, BA cabin crew join fray. In: ReutersFeb 22nd 2010. URL: http://www.reuters.com/

[6] Gittell J H. The Southwest Airlines Way.McGraw-Hill, 2005. ISBN: 0-07-139683-7

[7] Krammer P, Scholz D. ALOHA - Air-craft Design for Low-Cost Ground Handling.In: mobiles, 35 (2009/2010). HAW Ham-burg, Department of Automotive and Aero-nautical Engineering, pp 60-63, 2009. URL:http://ALOHA.ProfScholz.de

[8] Köhler W, Schachtel G, Voleske P. Biostatistik.4th edition, Springer, 2007. ISBN: 978-3-540-37710-8

[9] Heinze W. Ein Beitrag zur quantitativen Anal-yse der technischen und wirtschaftlichen Ausle-gungsgrenzen verschiedener Flugzeugkonzeptefür den Transport großer Nutzlasten. ZLRReport 94-01, Technical Universität Braun-schweig, 1994.

[10] Heinze W, Österheld C M, Horst P. Multidiszi-plinäres Flugzeugentwurfsverfahren PrADO –Programmentwurf und Anwendung im Rah-men von Flugzeugkonzeptstudien. In: GermanAerospace Congress 2001, Hamburg, DGLR-2001-194, 2001.

[11] Werner-Westphal C, Heinze W, Horst P. Multi-disciplinary Integrated Preliminary Design Ap-plied to Future Green Aircraft Configurations.In: 45th AIAA Aerospace Sciences Meeting andExhibit, Reno, Nevada, AIAA 2007-655, 2007.

[12] Association of European Airlines. Short-Medium Range Aircraft AEA Requirements.Brussels, G(T)5656, 1989.

[13] Schumacher A. Optimierung mechanischerStrukturen. Springer, 2005. ISBN: 3-540-21887-4

[14] Airbus S.A.S. A320 Airplane Characteristicsfor Airport Planning AC. Blagnac Cedex,France, Airbus S.A.S., Issue: Jul 1995, Ref: D -AC

[15] AIDA Development GmbH. Foldable Passen-ger Seat: Feasibility, Design and Mock Up.Corporate Information. URL: http://www.aida-development.de (2010-06-30)

[16] European Union. European Research ProjectClean Sky JTI. URL: http://www.cleansky.eu/(2010-06-30)

[17] Airport Research Center. CAST ComprehensiveAirport Simulation Tool - CAST Overview.Aachen, 2010, URL: http://www.airport-consultants.com/ (2010-03-15)

[18] U.S. Department of Labor, Bureau ofLabor Statistics. Consumer Price In-dex – All Urban Consumers, U.S. CityAverage. Washington, 2010, URL:ftp://ftp.bls.gov/pub/special.requests/cpi/cpiai.txt (2010-03-22).

[19] Beyer H T. Online-Lehrbuch – Allgemeine Be-triebswirtschaftslehre. Erlangen, University ofErlangen, Institut for Economics, Lecture notes,2004, URL: http://www.economics.phil.uni-erlangen.de/bwl/ (2010-03-28)

[20] Crönertz O. Prozessorientierte Kalkulation vonFlughafenleistungen. Schwerpunkt: Boden-abfertigungsdienste von Passagierflugzeugen.Saarbrücken, VDM Verlag Dr. Müller, 2008,ISBN: 978-3-8364-8460-2

[21] Krammer P, Rico Sánchez D, Scholz D. CostEstimation and Statistical Analysis of GroundHandling Process. Aero - Aircraft Designand Systems Group, Hamburg University ofApplied Sciences, 2010. Technical Report(ALOHA_FinalReport_Cost_and_Statistical_Analysis_2010-03-31.pdf)

[22] Sanz de Vicente S. Ground Handling Simulationwith CAST. Aero - Aircraft Design and SystemsGroup, Hamburg University of Applied Sci-ences, Master Thesis, 2010. Publication pend-ing

Copyright Statement

The authors confirm that they, and/or their company ororganization, hold copyright on all of the original ma-terial included in this paper. The authors also confirmthat they have obtained permission, from the copy-right holder of any third party material included in thispaper, to publish it as part of their paper. The authorsconfirm that they give permission, or have obtainedpermission from the copyright holder of this paper, forthe publication and distribution of this paper as part ofthe ICAS2010 proceedings or as individual off-printsfrom the proceedings.

13


2 3 4 5 6 7 8 9 10 11 120

0.05

0.1

0.15

0.2

0.25

disembarking time [min]

De

nsity P

rob

ab

ility

disembarking time

Normal

Lognormal(selected)

50 100 150 2002

4

6

8

10

12

no. of PAX disembarking

y = 0.046*x

data 1linear

4 6 8 10 12 14 16 180

0.02

0.04

0.14

0.16

embarking time [min]

De

nsity P

rob

ab

ility

embarking time

Normal(selected)Lognormal

40 60 80 100 120 140 160 1804

6

8

10

12

14

16

18

no. of PAX embarking

data 1

0.0

0.1

0.2

0.9

1.0

Cu

mu

lative

Pro

ba

bili

ty

0.0

0.2

0.4

0.6

0.8

1.0

Cu

mu

lativ

e P

rob

ab

ility

Disembarking Embarking

dis

em

ba

rkin

g t

ime

[m

in]

em

ba

rkin

g t

ime

[m

in]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

airline business model 1 1.000 0.163 -0.177 -0.175 -0.494 0.242 0.828 -0.082 -0.348 -0.286 -0.485 -0.153 -0.072 -0.078 0.047

A/C model 2 1.000 0.170 0.286 0.627 0.423 0.211 -0.063 0.319 0.326 -0.424 0.039 0.146 0.527 0.397

previous stage length 3 1.000 0.689 0.348 -0.059 -0.269 0.351 0.317 0.221 0.112 -0.109 -0.045 0.340 N/A

subsequent stage length 4 1.000 0.425 -0.009 -0.246 0.404 0.306 0.364 0.034 -0.161 -0.068 N/A -0.075

no. of A/C seats 5 1.000 -0.163 -0.541 0.301 0.749 0.666 0.185 -0.054 0.053 0.512 0.222

aisle width 6 1.000 0.349 -0.419 -0.404 -0.448 -0.575 0.332 0.233 0.083 0.130

seat pitch 7 1.000 -0.382 -0.447 -0.356 -0.478 -0.097 -0.070 -0.111 0.085

cabin class layout 8 1.000 0.335 0.262 0.293 -0.297 -0.159 -0.033 -0.283

no. of disembarking PAX 9 1.000 0.593 0.408 -0.274 -0.250 0.568 N/A

no. of embarking PAX 10 1.000 0.237 -0.221 -0.120 N/A 0.296

dis/embarking thr. 1/2 doors 11 1.000 -0.414 -0.544 -0.238 -0.354

time for positioning GSE 12 1.000 0.588 0.115 N/A

time for removing GSE 13 1.000 N/A 0.265

disembarking time 14 1.000 0.308

embarking time 15 1.000

Pearson product-moment correlation coefficients (PMCC):

Fig. 1 Example results of turnaround process analyses [21]: disembarking (left) versus embarking (right) withanalyses of linear regression (top), probability density distribution (middle) and PMCC (bottom)

14


0

25

50

75

100

0 2 4 6 8 10flight time [h]

rela

tive

daily

utiliz

ation

[%]

ceiling of with = 0.5 hU td,f,rel a

U td,f,rel awith = 0.5 h

floor of U td,f,rel awith = 0.5 h

U td,f,rel awith = 0.33 h

Fig. 2 Ud, f ,rel over t f at different ta with Ad = 18 h,tt = 10 min

0

10

20

30

40

0 1 2 3flight time [h]

for 30 min standard turnarounds

ceiling of 30 min standard turnarounds

for 20 min standard turnarounds

reduction in r

equired turn

aro

und tim

e [m

in]

reduction in necessaryta

Fig. 3 Necessary reduction in ta as a function of t f toincrease Ud, f with Ad = 18 h, tt = 10 min

Table 1 Results of A320 redesign to validate the de-sign method of PrADOParameter Airbus[14] PrADO Dev.operating empty mass 41244 kg 41966 kg 2%fuel mass (design range= 3273 km)

13157 kg 13401 kg 2%

max. take-off mass 73500 kg 74466 kg 1%cruise lift-to-drag ratio - 18.1 -required take-off fieldlength FAR

2090 m 2371 m 13%

required landing fieldlength FAR

1750 m 1603 m -8%

approach speed 72 m/s 82 m/s 13%

Table 2 Requirements for aircraft design proposals(including chosen values that are subject to the vali-dation of the baseline aircraft)Parameter Valuetransport mission- design range 3273 km- payload : 19099 kg- passenger (BC / EC) 12 / 138- cargo mass 5674 kgrange for flight with max. fuel 5277 kmmax. acceptable take-off field length(MTOW, 0 km, ISA)

< 2400 m

max. acceptable landing fieldlength(MLW, 0 km, ISA)

< 1650 m

max. acceptable approach speed(MLW, 0 km, ISA)

< 82 m/s

fuel reserves- range for diversion (reserve fuel)FAR Part 121 (domestic)

370.4 km

- loiter FAR Part 121 (domestic) 0.5 hfreight density- baggage 200 kg/m3

- cargo 180 kg/m3

Table 3 Results of investigated aircraft designsParameter Shoulder

Wing -single aisle

ShoulderWing -

twin aislefuselage geometry- length 40.6 m 40.6 m- max. width/height 3.95 m/4.14 m 4.43 m/4.14 mwing geometry- reference area 130 m2 133 m2

- aspect ratio 10.04 10.04static thrust 2 x 121 kN 2 x 121 kNcruise (begin)- altitude 11.28 km 11.28 km- Mach number 0.75 0.75- trim angle -9.5◦. . . 1.6◦ -10.7◦. . . 0.6◦

- L/D (trimmed) 18.37 18.73oper. empty mass 48295 kg 50009 kg- fuselage 11746 kg 12516 kg- wing 10153 kg 10561 kg- propulsion 7955 kg 7955 kgfuel mass (referencemission: 3273 km),incl. reserves

13401 kg 13401 kg

max. take-off mass 80795 kg 82509 kgmax. landing mass 71011 kg 72725 kgmax. loadable fuelmass

21212 kg 24897 kg

FAR take-off fieldlength

2356 m 2400 m

FAR landing fieldlength

1627 m 1631 m

approach speed 81.00 m/s 81.88 m/s

15


9000

9500

10000

10500

11000

11500

5000 7000 9000 11000 13000 15000 17000

-20

-10

0

10

20

30

Pareto-optimal boundary of wing mass v. fuel mass

wing area

lift-to-drag ratio

operating empty mass

wing mass [kg]

fuel m

ass [kg]

para

mete

r variation [%

]w

ith b

aselin

e w

ing a

s a

datu

m

baseline wing (fuselage mass = 9273 kg)

shoulder wing(fuselage mass = 11678kg)

final shoulderwing configuration

design variables:

variable min max

AR

S

6 15

0.2 1.080 160

l

ref

Fig. 4 Pareto optimal boundary of min. wing mass v. min. fuel mass of the shoulder wing (red curve) vs. thebaseline wing (red cross); note that the top left point could even be at higher fuel masses because the optimizationat this point was restricted by the preselected aspect ratio lower limit of AR = 6.

time [min]

0:00:00 0:05:00 0:10:00 0:15:00 0:20:00 0:25:00 0:30:00 0:35:00

Push Back

Ground Power Unit

Loading AFT

Loading FWD

Offloading FWD

Offloading AFT

Embarking

Waste Water Service

Potable Water Service

Cleaning

Catering

Refuelling

DisembarkingBeginn

Positioning

Connecting/Preparing

Process

Disconnecting/Postpositioning

Removing

Changing

Fig. 5 Turnaround Gantt chart of ground handling scenario I of the baseline aircraft [22]: process bars scaledaccording to regression analyses, lines represent standard deviations out of statistical evaluations

16


Baseline AircraftShoulder Wing Aircraft

(single aisle)

Shoulder Wing Aircraft(twin aisle)

cabin layout incorporatingfoldable passenger seatsnext to the aisle

fuselage cross section(superellipse)

example of a two class cabin layout for twin aisle configuration

cabin layoutincorporatingfoldablepassengerseats next tothe aisle

side and front view of shoulder wingaircraft (single aisle)

continuous cargo compartment

optimum aircraft CG positioning byoptimum container positioning through

ball mats and installed power driveunits on cargo floor panels

Fig. 6 PrADO 3D visualization using Tecplot of baseline aircraft (top left), single aisle shoulder wing aircraft(top right), and twin aisle shoulder wing aircraft (not further investigated in terms of ground handling).

17


Table 4 Definition of ground handling scenarios: conventional vs. low cost airline business model / terminal vs.remote apron positionScenario I II III IVairline business model conventional low costno. of passengers 67 % passenger load factor 83 % passenger load factorfuel according to DOC mission (range = 500 nm)catering two catering trucks: 1 AFT, 1 FWD one catering truck: 1 AFTpotable water service 100 l n/a 100 l n/awaster water service 80 l n/a 80 l n/aparking position terminal remote apron terminal remote aproncargo (type and amount) 4 ULDs

(3 AFT, 1 FWD)4 ULDs

(3 AFT, 1 FWD)100 bags

(bulk cargo)100 bags

(bulk cargo)ground power from PBB1 from GPU from PBB1 from GPUcleaning yes no yes nopush back towbarless n/a (remote apron) towbar n/a (remote apron)1 PBB = passenger boarding bridge

Fig. 7 Turnaround simulation with CAST Ground Handling at different time stamps of the baseline aircraft (3Dgeometry out of PrADO): ground handling scenario I at the top (left 04:45, right 10:57), scenario III at the bottom(left 05:45, right 12:09); left: PAX disembarking; right: all PAX disembarked. Turnaround simulation of adaptedaircraft designs is still in progress.

18


00:00

05:00

10:00

15:00

20:00

25:00

30:00

I II III IVGround Handling ScenarioGround Handling Scenario

Tu

rna

rou

nd

Co

sts

[U

S $

]

Tu

rna

rou

nd

Tim

e [

min

]

0

200

400

600

800

1000

I II III IV

-1.64 %

3.70 %

-1.66 %

3.63 %

4.05 %

-2.94 %

13.61 %

2.26 %

16.66 %

-3.77 %

4.43 %0 %

-1.40 %

3.28 %3.95 %

0 %

-1.42 %

4.31 %

10.00 %

-3.18 %

2.04 %

-4.02 %

4.22 %

9.45 % 9.52 %

-10

-5

0

5

10

15

20

-26.36 %

-20.45 %

14.29 %

-3.99 %

18.72 %

1.64 %

-42.57 %

-14.99 %

0 %

-1.29 %

3.52 %

4.24 %

-26.06 %

-18.10 %

-3.76 %

4.22 %

-28.41 %

-19.82 %

-1.52 %

4.03 %

9.44 %

-50

-40

-30

-20

-10

0

10

20

30

8.06 %6.21 % 6.85 %

8.04 %

-3.85 %

1.49 %

15.08 %

28.26 %

18.78 %

9.33 %8.50 %

-10

-5

0

5

10

15

20

25

30

35

fuselagelength

wingreference

area

aspectratio

staticthrust

cruiseMach

number

L/Dtrimmed

operatingemptymass

fuselagemass

propulsionmass

max.take-offmass

Baseline:Reference Aircraft

wingmass

I II III IV I II III IV

I II III IV

I II III IV


with Ground Handling Scenarios I, II, III, and IV

reductionin overall

turnaroundtime

reductionin totalground

handlingcosts

impact on DOC under the premiseof masses and performance

identical to the baseline aircraft

predicted turnaroundimprovements

change inrelativeannual

utilization

change inDOC per

seat-kilometre


utilization

change inDOC per

seat-kilometre

impact on DOC by applyingproposed single-aisleshoulder wing aircraft

change in fuelmass neededfor DOC flight

DOC flight range = 500 nm

Pa

ram

ete

r C

ha

ng

e [

%]

Pa

ram

ete

r C

ha

ng

e [

%]

I II III IV

I II III IV

I II III IV I II III IV I II III IV

I II III IV

DOC flight range = 300 nm DOC flight range = 700 nm





Baseline:Reference Aircraft


utilization


utilization


utilization


utilization



change inDOC per

seat-kilometre

change inDOC per

seat-kilometre

change inDOC per

seat-kilometre

change inDOC per

seat-kilometre

with basic DOC values of fuel: 0.68 US $ / kg, landing fee: 2463 US $, navigation fee: 781 US $, daily availability 16 h, annual availability 3750 h for all calculations

Pa

ram

ete

r C

ha

ng

e [

%]




Fig. 8 Assessment of turnaround improvements in general and by applying the proposed aircraft design

19

aircraft design for low cost ground … design for low cost ground handling - the final results of...

Documents