Aircraft Engineering and Aerospace Technology

Development of a civil aircraft dispatch reliability prediction methodology

Mansour Bineid J.P. Fielding
Article information:
To cite this document:
Mansour Bineid J.P. Fielding, (2003),"Development of a civil aircraft dispatch reliability prediction methodology", Aircraft
Engineering and Aerospace Technology, Vol. 75 Iss 6 pp. 588 - 594
Permanent link to this document:
http://dx.doi.org/10.1108/00022660310503066
Downloaded on: 22 October 2014, At: 05:18 (PT)
References: this document contains references to 3 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 668 times since 2006*
Users who downloaded this article also downloaded:
(1979),"Retrofit", Aircraft Engineering and Aerospace Technology, Vol. 51 Iss 10 pp. 17-18
(1985),"Sidestick Control Evaluation Completed for Tilt#Rotor Aircraft", Aircraft Engineering and Aerospace Technology, Vol.
Downloaded by Ms Maria Makrygianni At 05:18 22 October 2014 (PT)

57 Iss 11 pp. 13-13

(1985),"‘Samson Pod’ Gives Hercules New Electronics Capability", Aircraft Engineering and Aerospace Technology, Vol. 57
Iss 6 pp. 18-18

Access to this document was granted through an Emerald subscription provided by Token:Reviewer:A97179E8-14E2-4D66-
A290-0BEBBCB53E8E:
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service
information about how to choose which publication to write for and submission guidelines are available for all. Please
visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of
more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online
products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication
Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.


Paper
Development of a civil
aircraft dispatch
reliability prediction
methodology
Mansour Bineid and
J.P. Fielding
The authors
Mansour Bineid and J.P. Fielding are based at
Aerospace Engineering Group, School of Engineering,
Downloaded by Ms Maria Makrygianni At 05:18 22 October 2014 (PT)
Cranfield University, Bedford, UK
Keywords
Aircraft, Design and development, Maintenance reliability
Abstract
This paper explains the development of a dispatch
reliability prediction method for passenger aircraft.
The method provides an objective means of assessing the
delay rate for aircraft systems in the very early stages of
aircraft design, leading to the minimisation of subsequent
corrective actions, and thus reducing operating cost.
The method encompasses two models: one each for
short-haul and long-haul aircraft. It uses delay rate data
and design/performance parameters to derive statistical
equations, one set for each aircraft system. These
equations relate the delay rate for a particular system to
selected design/performance parameters. The models give
high predictability and can be applied at the conceptual
design phase, using the minimum amount of input data.
Electronic access
The Emerald Research Register for this journal is
available at
http://www.emeraldinsight.com/researchregister
The current issue and full text archive of this journal is
available at
http://www.emeraldinsight.com/0002-2667.htm
Aircraft Engineering and Aerospace Technology
Volume 75 · Number 6 · 2003 · pp. 588–594
q MCB UP Limited · ISSN 0002-2667
DOI 10.1108/00022660310503066
588
1. Introduction
1.1 The need for a methodology
It is well known that some 65 per cent or more
of the life-cycle costs of a design are locked-in
by the end of the conceptual phase of product
development (Fabrycky and Blanchard,
1991). This necessitates getting the design
right from the start. Reliability is one of the
design features that need to be considered
from the outset, and this is particularly true
for aircraft dispatch reliability[1].
This leads to the need for developing the
aircraft dispatch reliability prediction method,
which could be used for:
.
establishing realistic baseline reliability
requirements for the planned aircraft;
.
evaluating the reliability characteristics of
alternative aircraft system designs during
the conceptual phase; and
.
evaluating competing designs against
clear numerical targets.
1.2 Requirements
Although some aircraft dispatch reliability
prediction models exist (Fielding, 1999),
there is a need for a model that uses
up-to-date aircraft data, to provide design
engineers with a quantitative system reliability
objective at the beginning of the design effort.
The developed model should provide
mathematical expression to predict the delay
rate for every aircraft system in accordance
with ATA100[2] chapters; this, in turn will
allow prediction of the dispatch reliability for
the whole aircraft, during the conceptual
design phase.
Therefore, the objectives of the method
are to:
.
perform reliability assessments of a wide
spectrum of aircraft design concepts;
.
be responsive to variations in aircraft
design features; and
.
be capable of providing representative
reliability values, based on historical data.
The model should relate aircraft delay rates
(i.e. reliability design parameters) to aircraft
design/performance parameters.
The mathematical expressions should be
developed by statistical analysis of relation-
ships between delay rate and aircraft design/
performance parameters. Because the pre-
diction model is intended for use in the
conceptual design phase, the values of the
parameters needed in the prediction equation
should be available during that phase.
 Development of a civil aircraft Aircraft Engineering and Aerospace Technology
Mansour Bineid and J.P. Fielding Volume 75 · Number 6 · 2003 · 588–594

2. Prediction method development Table I Design/performance parameters

No. Design parameter Notation Units
The development of the prediction model
involved three major tasks: 1 Maximum aircraft takeoff weight MTW kg
(1) database development; 2 Total number of passengers NP –
(2) derivation of the prediction equations; 3 Maximum aircraft engine thrust Thr kN
and 4 Flight length FL hr
(3) validation of the model. 5 Number of engines NE –
6 Maximum aircraft length MAL m
2.1 Database development 7 Total ECS weight ECSW kg
There is very limited aircraft reliability 8 Total fuel capacity FC kg
information in the literature, and what is there
should be treated with caution (O’ Conner, them and the delay rates were examined.
1991). The development of a passenger The new calculated parameters are shown in
aircraft dispatch reliability method requires Table II.
the use of accurate and up-to-date
information. Therefore, new data to be used
2.3 Selection of the design/performance
in the development of this method had to be
parameters
obtained from current airlines and
Downloaded by Ms Maria Makrygianni At 05:18 22 October 2014 (PT)

Engineering judgment was applied to select

manufacturer sources.
the most suitable parameters that had logical
It was decided to select information for this
relationships with particular systems. By using
method from different manufacturers and
this concept, elimination of those parameters
operators. This makes the model general, and
having no association with a given system
not subject to a particular size, role or
could be achieved.
technology level.
The selected parameters were then
As large a number of aircraft as possible was
subjected to correlation analysis, which
used in the method, because this offers more
examined the relationship between the delay
representative aircraft delay rates.
rate and the selected parameter under
The selected aircraft were divided into two
analysis. Afterwards, parameters which had a
types, according to flight times. Three hours
strong association (high correlation) were
or less is considered to be a short-haul flight,
selected to be part of the prediction equation
while over 3 h is seen as long-haul.
for the system under discussion.

2.2 Aircraft design/performance

2.4 Data analysis
parameters database
Correlation coefficients between system
As the developed prediction method is
parameters and delay rates were calculated,
intended to be used during the conceptual
and the parameter with the highest correlation
design phase, the design/performance
coefficient selected. In cases when
parameters that are employed should be
the correlation coefficient was low (i.e. the
the ones available at that time. The
correlation between the delay rate and the
design/performance database included eight
selected parameter was weak), the natural log
parameters for each of the aircraft in the
or exponential of the delay rate was used and
model, which were collected from different
the correlation coefficient was then calculated.
sources. The sources included:
.
data from the aircraft and engine
3. Results
companies;
.
the internet sites of aircraft Following data analysis, a graphical
manufacturers; representation was developed of the
.
Jane’s: all the world’s aircraft; relationship between the selected parameter
.
flight international magazine; and which had the highest correlation coefficient
.
aircraft design books.
The design/performance parameters are Table II New calculated parameters
presented in Table I. No. New calculated parameters Notation Units
For the purpose of obtaining the most
suitable relationship, some new parameters 1 Maximum aircraft weight/flight length MTW/FL kg/hr
were added and the relationship between 2 Total number of passengers/flight length NP/FL –

589
 Development of a civil aircraft
Mansour Bineid and J.P. Fielding
and delay rate. As an example, the
communications system (ATA 23) is shown in
Figure 1. The delay rate equation for every
system was derived, and the prediction
equations for all the ATA systems were
produced for short-haul and long-haul,
aircraft.
The equations are presented in Tables III
and IV, respectively. The parameters used in
the equations are shown in Table V.
The developed delay rate equations can be
used to predict the delay rate in individual
systems. Thus, the dispatch reliability for the
whole aircraft can be calculated, using
equation (1), by subtracting the resulting total
delay rates from 100. Hence,
Dispatch reliability ðper centÞ
Downloaded by Ms Maria Makrygianni At 05:18 22 October 2014 (PT)
¼ 100 2 delay rate
where the delay rate is the total aircraft delay
rate, counting delays over 15 min. This can be
calculated by carrying out a summation of
the predicted delay rate for all the aircraft
systems.
The developed method also provides
another way of predicting the whole aircraft
dispatch reliability. By using only the total
dispatch reliability figure, two prediction
equations have been produced by collecting
data for many aircrafts and analysing their
relationships to different aircrafts
design/performance parameters.
This resulted in equation (2) for short-haul
aircraft and equation (3) for long-haul
aircraft. These predict whole aircraft dispatch
reliability without using the individual system
aircraft delay rates.
Figure 1 Delay rate vs no. of passengers
Aircraft Engineering and Aerospace Technology
Volume 75 · Number 6 · 2003 · 588–594
Dispatch reliability ¼ 1:8893 ln ðThrÞ
þ 87:954 ð2Þ
Dispatch reliability ¼ 20:0066 ðNPÞ
þ 100:18 ð3Þ
4. Validation
For the validation process, delay rate data
were obtained for an aircraft that had not
been used in the method derivation.
The results for the short-haul aircraft are
shown in Table VI.
They show the high level of predictability of
the method for individual aircraft systems,
ð1Þ and also for the total aircraft delay rate.
To predict the whole aircraft dispatch
reliability without using the aircraft systems
delay rate, equation (2) was used.
The resulting predicted dispatch reliability
was 98.93, while the actual dispatch reliability
was 99.01.
The error percentage is a means of
measuring the predictability, which indicate
high predictability for the low error and low
predictability if the error is high. It is defined
as the:
predicted delay rate 2 actual delay rate
actual delay rate
£ 100
The error for the whole aircraft dispatch
reliability was 0.08 per cent, which shows high
predictability.
590
 Development of a civil aircraft Aircraft Engineering and Aerospace Technology
Mansour Bineid and J.P. Fielding Volume 75 · Number 6 · 2003 · 588–594

The same procedure was performed Table V Aircraft parameters

for long-haul aircraft and resulting
CS Cruise speed
ATA delay rate comparison is shown in
Table VII. MTW Maximum take off weight
MTW/FL Maximum take off weight/flight length
Table III Short-haul delay rate prediction equations NP/FL Number of passengers/flight length
NP Number of passengers
ATA System Delay rate prediction equations
FL Flight length
21 Air conditioning ln(DR) ¼ 0.5479 ln(NP/FL) 2 5.3721 Thr Aircraft thrust
22 Auto pilot DR ¼ 2 0.0063(FL) + 0.0318
23 Communications DR ¼ 0.0002(NP) 2 0.0449
24 Electric power DR ¼ 2 0.0003(NP/FL) + 0.0668 Table VI Predicted and actual delay rates for short-haul
25 Equipment DR ¼ 0.0186(FL) + 0.008 aircraft
26 Fire protection DR ¼ 2 0.0001(Thr) + 0.0559 Predicted delay Actual delay
27 Flight control DR ¼ 2 0.0008(NP/FL) + 0.1937 ATA rate rate
28 Fuel ln(DR) ¼ 2 0.0161(NP/FL) 2 2.1535
21 0.05 0.04
29 Hydraulic power ln(DR) ¼ 0.2265 ln(FL) 2 2.5852
22 0.01 0.01
30 Ice protection DR ¼ 2 1 £ 102 6(MTW/FL) + 0.0892
23 0.01 0.00
31 Instruments DR ¼ 2 3 £ 102 7(MTW) + 0.054
Downloaded by Ms Maria Makrygianni At 05:18 22 October 2014 (PT)

24 0.05 0.04
32 Landing gear DR ¼ 0.0012(NP) 2 0.3128
25 0.06 0.03
33 Lights DR ¼ 0.0001(NP) + 0.0003
26 0.02 0.00
34 Navigation DR ¼ 2 0.0005(NP) + 0.1548
27 0.14 0.09
35 Oxygen DR ¼ 2 6 £ 102 7(MTW/FL) + 0.0303
28 0.04 0.03
36 Pneumatics DR ¼ 0.0013(CS) 2 0.531
29 0.10 0.09
38 Water waste DR ¼ 3 £ 102 7(MTW) 2 0.0067
30 0.05 0.04
49 APU DR ¼ 2 0.0002(NP) + 0.0702
31 0.02 0.02
51 Structure DR ¼ 0.0002(Thr) 2 0.0725
32 0.09 0.13
52 Doors DR ¼ 0.0006(NP) 2 0.1351
33 0.02 0.01
71-80 Powerplant ln(DR) ¼ 2 0.0063(NP/FL) 2 1.1094
34 0.06 0.06
35 0.01 0.00
Table IV Long-haul delay rate prediction equations 36 0.07 0.09
38 0.02 0.02
ATA System Delay rate prediction equations
49 0.03 0.03
21 Air conditioning DR ¼ 20.007 (FL) + 0.0963 51 0.01 0.00
22 Auto pilot DR ¼ 0.0003(NP/FL) 2 0.0165 52 0.02 0.02
23 Communications DR ¼ 5 £ 102 7(MTW) 2 0.058 71-80 0.22 0.24
24 Electric power DR ¼ 20.0216(FL) + 0.1818 Total delay rate 1.09 0.99
25 Equipment DR ¼ 0.0019(NP/FL) 2 0.0676
26 Fire protection DR ¼ 3 £ 102 7(MTW) 2 0.0191
27 Flight control DR ¼ 0.0029 (NP/FL) 2 0.0772
In order to predict the whole aircraft dispatch
28 Fuel DR ¼ 0.0005(NP) 2 0.0549 reliability without using the individual aircraft
29 Hydraulic power DR ¼ 0.0003 (NP) + 0.0203 systems delay rate, equation (3) was used.
30 Ice protection DR ¼ 2 £ 102 6(MTW/FL) 2 0.0457 The resulting predicted dispatch reliability
31 Instruments DR ¼ 9 £ 102 8(FL) + 0.0073 was 98.93, while the actual dispatch reliability
32 Landing gear DR ¼ 0.0031(NP/Fl) 2 0.0455 was 98.92. The error percentage was 0.01,
33 Lights DR ¼ 1 £ 102 6(FL) 2 0.0237 which shows a very high level of predictability.
34 Navigation DR ¼ 0.0043(NP/FL) 2 0.1511
35 Oxygen DR ¼ 0.0001(NP/FL) 2 0.006
36 Pneumatics DR ¼ 20.0227(FL) + 0.2399
5. Discussion
38 Water waste DR ¼ 20.005(FL) + 0.0716
49 APU DR ¼ 20.0154(FL) + 0.1315 (1) The development of a dispatch reliability
51 Structure DR ¼ 0.0014(NP/FL) 2 0.0728 method for passenger aircraft involved an
52 Doors DR ¼ 0.0005(NP) 2 0.0947
enormous amount of data analysis. This
56 Windows DR ¼ 3 £ 102 6(NP)
was accomplished using the programming
57 Wings DR ¼ 21 £ 102 5(CS) + 0.0051
capability of the EXCEL spread sheet.
71-80 Powerplant DR ¼ 0.0145(NP/FL) 2 0.5566
A total of 500 correlations between delay
591
 Development of a civil aircraft Aircraft Engineering and Aerospace Technology
Mansour Bineid and J.P. Fielding Volume 75 · Number 6 · 2003 · 588–594

Table VII Predicted and actual delay rates for long-haul correlation coefficients. The reasons are
aircraft as follows.
Predicted delay Actual delay
.
The aircraft system individual delay
ATA rate rate rate equations may have small
percentages of error which
21 0.06 0.090
accumulate. This decreases the
22 0.00 0.004
accuracy of this part of the
23 0.02 0.019
methodology.
24 0.08 0.076
25 0.04 0.070
.
Differences of the delay rate input
26 0.03 0.011 information between operators may
27 0.09 0.110 also degrade the accuracy of the
28 0.08 0.026 aircraft system individual delay rates
29 0.10 0.064 equations.
30 0.02 0.016 (6) The accuracy of equations (2) and (3) to
31 0.01 0.020 predict the whole aircraft dispatch
32 0.13 0.220 reliability for short- and long-haul aircraft
33 0.00 0.035 is also very high (the error percentages
34 0.10 0.161 were 0.08 and 0.01 per cent,
Downloaded by Ms Maria Makrygianni At 05:18 22 October 2014 (PT)

35 0.00 0.000 respectively).

36 0.13
38 0.05
49 0.06
51 0.01
52 0.04
56 0.00
57 0.00
71-80 0.27
Total delay rate 1.30
rate data and aircraft performance/design
parameters were performed, and more
than 350 graphs plotted.
(2) The data collection was a very difficult
and time-consuming task.
(3) All the aircraft used in the derivation of
this method were in service for more than
3 years, that is to say, they had reached
maturity.
(4) Data concerning the aircraft delay rates
that were used for the production of the
method in this research were the average
for 1 year’s operation; this was to
overcome seasonal effects. The delay rate
data for a particular aircraft type were
calculated from different operators’ data,
where possible, in order to eliminate the
effect of different styles of operation.
(5) The predictability for most of the
individual system equations is very high,
but when summed for the whole aircraft
leads to errors of some 10 per cent.
The whole aircraft dispatch reliability
equation, however, is very accurate;
this is shown by the very high
0.142 (7) Calculating the ATA average delay rate
0.086 for all the aircraft reveals many interesting
0.061 points, they are as follows.
0.008 .
The ATA chapter 71 – 80 (power
0.026 plant) has the highest delay rate for
0.004 both short- and long-haul aircraft.
0.000
This can be seen graphically in
0.268
Figures 2 and 3. This is probably due
1.52
to the complexity, hard operating
environment; high stress levels and
continuous demand on this system.
.
The system which shows the next
highest delay rate is the ATA 32
(landing gear) for the short-haul
aircraft. This is probably due to the
high number of landings per day
performed by short-haul aircraft.
.
Flight control and hydraulic systems
for short-haul aircraft shared the third
highest delay rate percentage. This is
due to the characteristics of the
short-haul flight, which impose more
work load on the flight controls per
flying hour. The same can be said
about the hydraulic system, which
provides the power needed by the
flight controls.
.
For long-haul aircraft the second
highest delay rate is shared by more
than one system. They are the same
systems as for short-haul (landing gear
and hydraulic systems) plus the fuel
system which is to be expected for
long-haul airplanes, because of the size
and complexity of this system.
592
 Development of a civil aircraft
Mansour Bineid and J.P. Fielding
Figure 2 Average delay rates for short-haul aircraft (see Table IV for chapter title)
Downloaded by Ms Maria Makrygianni At 05:18 22 October 2014 (PT)
Figure 3 Average delay rates for long-haul aircraft
6. Conclusions and recommendations
(1) The developed methodology provides a
technique to predict the aircraft dispatch
reliability during the very early design
stage, with good accuracy.
(2) Using more delay rates data will improve
the derivation of the prediction equations
since it will reduce information scatter.
(3) Each prediction equation should be used
for the intended aircraft category
(long-/short-haul).
(4) A similar prediction methodology could
be developed for military aircraft, taking
into account the use of different
parameters.
(5) Although the help from the aerospace
community was vital to the
593
Aircraft Engineering and Aerospace Technology
Volume 75 · Number 6 · 2003 · 588–594
accomplishment of this methodology and
it is appreciated, there was reluctant
attitude from some operators.
(6) It is suggested that delay rates per aircraft
system, and dispatch reliability data
should be available in the public domain.
The aircraft operators and manufacturer
should feed this data into one organization
acting as central data base. That would
make it available to the researcher,
specialist and to the public. The origins of
the data could be treated as confidential,
to protect commercial interests.
(7) The analysis of the ATA average delay
rate for all the aircraft showed that
dispatch reliability improvement can be
achieved by placing more emphasis on
 Development of a civil aircraft
Mansour Bineid and J.P. Fielding
the design, operation and maintenance
of the power plant, landing gear, flight
control, hydraulic systems and fuel
systems since they are the systems that
cause most of the delays.
(8) Modifying some of the operation and
maintenance procedures for those
systems should enhance their dispatch
reliability.
(9) The validity of this method is restricted to
the aircraft system-level predictions.
More detailed dispatch reliability
predictions, such as aircraft sub-systems
and LRU, are not possible by this
method, but should be developed.
Notes
Downloaded by Ms Maria Makrygianni At 05:18 22 October 2014 (PT)
1 Dispatch reliability is defined as “the percentage of
scheduled flights, which departs without making a
594
Aircraft Engineering and Aerospace Technology
Volume 75 · Number 6 · 2003 · 588–594
mechanical delay of more than 15 min or
cancellation”. The delay rate definition is “the
percentage of scheduled departures which are more
than 15 min late or are cancelled”. From the above
two definitions, it can be seen that the dispatch
reliability is the complement of the delay rate and
vice versa.
2 ATA100 is a document produced by Air Transport
Association of America, it is commonly used by the
aviation community to categorise aircraft systems
into a 100 aircraft systems hierarchy.
References
Fabrycky, W.J. and Blanchard, B.S. (1991), Life-Cycle Cost
and Economic Analysis, Prentice-Hall, London.
Fielding, J.P. (1999), Introduction to Aircraft Design,
Cambridge University Press, Cambridge.
O’ Conner, P.D.T. (1991), Practical Reliability Engineering,
3rd ed., Wiley, New York, NY.
 This article has been cited by:

1. Rogério S. Rodrigues, Pedro Paulo Balestrassi, Anderson P. Paiva, Alberto Garcia-Diaz, Fabricio J. Pontes. 2012. Aircraft
interior failure pattern recognition utilizing text mining and neural networks. Journal of Intelligent Information Systems 38:3,
741-766. [CrossRef]
2. Feyza Gürbüz, Lale Özbakir, Hüseyin Yapici. 2009. Classification rule discovery for the aviation incidents resulted in fatality.
Knowledge-Based Systems 22:8, 622-632. [CrossRef]
3. Stephen Liddle, William CrowtherSystems and certification issues for active flow control systems for seperation control on
civil transport aircraft . [CrossRef]
Downloaded by Ms Maria Makrygianni At 05:18 22 October 2014 (PT)