Conference PaperPDF Available

Development and Simulation of a new Scheme for the Aircraft Cleaning Service

Authors:

Abstract and Figures

During the last decade with the increase of competition, airlines have set up schemes to lower costs. Their present profit margin has narrowed to the point of not being able to compete with companies whose business model is similar to the low-cost ones forcing them to explore novel ways of managing the available resources in order to keep competitive. One of the costs is the cleaning service generated by contracting this service and the delays that this operation can cause. The aim of this paper is to propose a new management system for scheduling the on board cleaning service, that lowers current costs, using tools such as modelling with coloured petri nets and simulation.
Content may be subject to copyright.
DEVELOPMENT AND SIMULATION OF A NEW SCHEME FOR THE
AIRCRAFT CLEANING SERVICE
Miguel Mújica (a), Mireia Soler (b), Idalia Flores (c)
(a) Amsterdam University of Applied Sciences
(b) Universitat Autonoma de Barcelona
(c)Universidad Nacional Autónoma de México
miguelantonio.mujica@uab.es, mireia.soler@campus.uab.es, idalia@unam.mx
ABSTRACT
During the last decade with the increase of competition,
airlines have set up schemes to lower costs. Their present
profit margin has narrowed to the point of not being able
to compete with companies whose business model is
similar to the low-cost ones forcing them to explore
novel ways of managing the available resources in order
to keep competitive.
One of the costs is the cleaning service generated by
contracting this service and the delays that this operation
can cause. The aim of this paper is to propose a new
management system for scheduling the on board cleaning
service, that lowers current costs, using tools such as
modelling with coloured petri nets and simulation.
Keywords: Simulation, coloured Petri nets, cleaning
services, aeronautics
1. INTRODUCTION
Years ago, airlines had enough capital to be able to
have their planes cleaned on each leg of a journey.
Moreover, plane tickets were much more expensive in
those days, with flying being luxury and longer stopover
times.
During the last decade, with the appearance on the scene
of low-cost airlines, airlines have set up schemes to lower
costs, as their present profit margin has narrowed to the
point of not being able to compete with such low prices
as these airlines offer for short and medium-haul flights.
One of the costs is the cleaning service and everything
involved with cleaning a plane, such as the cost of hiring
this service and the delays that this can cause.
The proposed system is based on modelling stopover
times, by simulating an airline’s flight schedule during a
working day.
2. ECONOMIC STRUCTURE OF AIRLINES.
An airline is an organization or company, devoted to
the transport of passengers, freight, mail and, in some
cases, life animals, using airplanes for a profit.
The economic structure of the airlines in existence at the
present time can be segmented as follows:
- Flag-carrying airlines: these are government-operated
airlines. They have a wide variety of planes for short- and
medium-and long-haul flights and tend to have a
monopoly on domestic flights.
- Traditional airlines: these are private companies for
passenger, freight or mail transport. They have a varied
fleet of planes and their routes can be short- and medium-
and long-haul. These are like the flag-carrying airlines
but with the difference that, in this case, governments are
not involved.
- Charter airlines: these are companies that transport
passengers but on an occasion basis, their method of
operation is to study the travel needs of a specific sector
of customers. They organize a group of passengers and
fit them up with a vacation package with the flight, hotel
and excursions included. They usually have a small fleet
of planes with capacity for approximately 180
passengers, per plane.
- Low-cost airlines: they supply the low-budget market
in exchange for eliminating passenger services. Their
strategy is to reduce operational and wage costs in order
to be able to give their customers very low and affordable
prices per route, thus achieving a broad customer base
that goes from people with high net worth to people with
a low level of purchasing power who would never been
in a position to buy a plane ticket.
3. STOP OVER TIMES AND MAINTENANCE.
The stopover of a plane is the temporary space
between consecutive flights when the plane is in the
airport. Depending on the type of company, the time and
space available, the plane’s stopover will be more or less
long (Basargan, 2004).
It is worth mentioning that every stopover takes a
different length of time, as all the flight schedules are
different. Moreover, it is impossible to homogenize the
times of all the ground handling processes when the
plane has already arrived at the airport.
The steps followed by a typical stopover of an aircraft
are:
- Prior preparation for boarding: the lines of
passengers are organized then all their hand luggage and
documentation is checked.
- The plane arrives at the parking stand.
- Block-In is performed.
- The passengers and bags disembark.
- The plane is fuelled.
-Whether there is a scheduled cleaning, the cleaning
team will proceed to clean the plane.
- When the last passenger leaves and the cleaning
services have finished, the passengers for the next flight
shall be boarded. Simultaneously the bags shall start to
be loaded on the new flight.
- During the boarding of passengers, the coordinator
shall deliver the necessary documentation to the captain.
- As soon as the plane is loaded with fuel, bags and
passengers, the doors are closed.
- The chocks are removed.
- The plane performs the taxiing towards the
corresponding runway for the take-off.
3.1. Maintenance of the Airplanes
There are three types of maintenance:
a) Daily check.
Inspect for obvious damage and check the general
conditions and security.
b) Minor maintenance.
A-check: performed every 500-800 flight hours, consists
in a general inspection of the systems, components and
structure of the aircraft and it can take 20-100 man-hours.
B-check: is done every 4-6 months, this is a slightly more
detailed check of components and systems and it can take
1-3 days.
C-check: is carried out every 15-21 months or after
specific flight-hours determined by the manufacturer,
this is a thorough inspection of the structures, the systems
and the inside and outside areas of the plane and it can
take 1-2 weeks.
c) Major maintenance.
Also called the “Heavy Maintenance Visit”. It covers the
full structural inspection program for the airplane. This
usually takes about two months and it should be done
every 5 years or 30,000 flying hours.
On the other hand, it sometimes happens that a plane
goes into AOG (Aircraft On Ground) which means that
the plane has a problem that is sufficiently serious to stop
it from making the next flight. In this case, the
maintenance team needs to go to the plane to solve the
fault.
4. OPERATION OF THE CLEANING SERVICE
IN AIRLINES
There are two ways of delivering the service:
a) Subcontracting a company. Every week, they
receive the stopovers schedule of each airplane and the
pair of origin and destination of the flights. With this
information, the cleaning service is scheduled, without
any modification throughout the week.
b) Performed by flight attendants. They are in
charge of cleaning the planes. The flight attendants have
signed an agreement in which they agree to do these
types of procedures and accept the conditions imposed
by the airline. The aim of this method is to reduce the
stopovers between one flight and another. This way the
plane spends more time in the air during the day.
5. CASE STUDY OF A SPANISH AIRLINE
A new scheme for the cleaning operations during
stopovers has been developed. The proposed scheme
uses information that has been provided by a Spanish
airline through a confidentiality agreement. We shall
refer to this airline, when applicable, as “the airline”. The
information of the schedule of one day has been used for
the model. The proposed schema is a particular one for
the case of the airline, but it can be extrapolated for the
case of other airlines in a very straightforward way.
5.1. Current cleaning activities
The following are the cleaning operations currently
under use by the airline.
Stopover cleaning.
This is the quickest way of cleaning and applies to
stopovers that last for more than 40 minutes, as well as
being the most common because, as the name says, it is
done during stopovers and it takes 8-14 minutes.
Extra cleaning.
This type of cleaning is unscheduled. The crew or
maintenance asks for some of the stopover cleaning jobs
to be done. There can be an unexpected use of the
temporary space of the stopover time. This type of
cleaning does not share all the characteristics of the
stopover cleaning. It only makes a required part of it.
However the service is charged as a stopover cleaning.
There were 137 extra cleanings during the month of
study.
Overnight cleaning.
This type of cleaning is done 4 or 5 times a week, when
the plane spends the night in an airport. As this cleaning
takes a long amount of time, it is done at night. It aims to
improving the plane’s level of disinfection and cleanses
places that cannot be reached during the stopover due to
the lack of time.
Deep cleaning.
This is a type of cleaning designed to totally disinfect and
clean the interior of the airplane. For this purpose, all the
seats and luggage compartments are dismantled. As it
takes too long, it is performed at night and once in a
month.
5.2. Impact of Cleaning Operations
The delays in the aviation industry are one of the
most important problems that the sector faces nowadays.
Due to the complexity and precedence relationships of
the aviation network one delay or primary delay caused
in an airport will propagate easily to the rest of the
network. Furthermore if more primary delays occur
during the day, at the end of the day the accumulated
delay would be sometimes huge (Jetzky 2009, Guest
2007). Every minute’s delay in the departure of a flight
Groups Length of Stopover Time Description
Group 1 Less than 41 minutes A very short stopover is contemplated
Group 2 Between 41 and 50 minutes. A short stopover is contemplated
Group 3 Between 51 and 60 minutes. A medium/long stopover is
contemplated
Group 4 Over 60 minutes. A long stopover is contemplated.
signifies an increase in the different rates that the airport
imposes on the airline.
The delays that directly affect the airline and a flight
are mainly because of:
• Handling
• Airport authority
• Auxiliary Services
• Safety
Meteorology
Cleaning service is a portion of the auxiliary services, in
which it generates 65% of the delays in scheduled flight
times for the airline.
The main characteristics of the current operation can be
defined as:
- It is an inflexible system that does not adapt to the
stopover times that airlines need under a fierce
competitive market.
- The number of cleanings can and must be reduced.
- It does not make much sense to charge for an extra
cleaning as if it were a stopover cleaning since the
cleaning performed is more superficial.
- The delays caused by the cleaning operation can
and must be reduced.
- More variables should be taken into account when
a cleaning is assigned, such as, the number of passengers
transported, number of previous cleaning among others.
- The current cleaning schedule is fixed and does not
admit the variability produced by a plane breaking down
or a request for an extra cleaning.
5.3. A novel operative schema for managing the
stopover times
There is a very high cost in having the plane standing due
mainly to the high tariffs demanded by the airports.
Moreover, if the stopover times during the day are
shortened, a plane can fly more hours, in other words the
useful life of the plane would be maximized. The more
hours a plane fly, the more flights it can do, the more
passengers it can transport and the less expenditure on
airport tariffs is incurred.
For these reasons, airlines seek to reduce the time their
planes spend in airports and to increase the number of
flights per plane.
However, shorter stopover times make the ground
handling of the plane all the harder.
To better manage the stopover time, a cleaning system
that fits with current needs must be designed.
New stopover times have been proposed and they are
organized into 4 groups that are presented in Table1
Table1: Length of Stopover Time
Using the proposed segmentation, a cleaning
management system has being designed for these new
stop over times.
The proposed model has 5 cleaning types:
- Cleaning 1. This type of cleaning has been
designed to give a basic and fast service, it takes 5-8
minutes. It shall be assigned in a very short stopover or
when the last cleaning is type 4 or 5.
- Cleaning 2. This type of cleaning gives the same
service as the stopover cleaning in the actual model. It
shall be assigned in a short and medium stopover or when
the last cleaning is type 4 or 5.
- Cleaning 3. This type has been designed to give a
good level of cleaning in medium and long stopovers,
and also to set back the cleaning number 4 and 5.
- Cleaning 4. This type of cleaning is just done once
a week during long stopovers, to give a better level of
disinfection and also set back the cleaning number 5.
- Cleaning 5. This type of cleaning is the same as
the deep cleaning in the actual model, yet it can be done
every month and a half.
6. DESCRIPTION OF THE CAUSAL MODEL
A causal model is proposed for evaluating the
cleaning operations, in which stopover times are grouped
according to the above division. The objective of the
causal model is to assess the validity of the proposed
schema while at the same time evaluate the magnitude of
savings that can be achieved.
The model was developed in the coloured petri net
formalism and tested using the CPNTools program.
6.1. Coloured Petri Nets
Coloured Petri Nets (CPN) is a simple yet powerful
modelling formalism which allows to properly modelling
discrete-event dynamic systems which present a
concurrent, asynchronous and parallel behaviour (Moore
et al. 1996, Jensen 1997, Christensen et al. 2001). CPN
can be graphically represented as a bipartite graph which
is composed of two types of nodes: the place nodes and
the transition nodes. The entities that flow in the model
are known as tokens and they have attributes known as
colours.
The formal definition is as follows (Jensen1997):
(,,,, , ,,,)CPN P T A N C G E I
Where
= { C1, C2, … , Cnc} represent the finite and
not-empty set of colours. They allow the
attribute specification of each modelled entity.
P = { P1, P2, … , Pnp} represent the finite set
of place nodes.
T = { T1, T2, … , Tnt} represent the set of
transition nodes such that P T =
which
normally are associated to activities in the real
system.
Place Colour Description
Airplanes airplane=product(ac*sa*p*h
*te1*te2*te3*te4*nt*a*q*n*s)
The initial state of this place
has 27 tokens with the
information of the first flight of
each airplane. This place will
keep track of the status of the
airplanes.
Next stopover new=product(ac*sa*sa2*p1*
h1*ne1*ne2*ne3)
This place has the flight
schedule information for each
airplane, except the first flight.
AOG aog
This place has 170 tokens to
generate the airplane-break-
down probability
Extra Cleaning le
This place has 252 tokens to
generate the request –extra-
cleaning probability.
Control y
This place controls the
activation of transition 1 or 2.
Decision airplanes1=product(p*h*te1
*te2*te3*te4*nt*s*a*q*n*b*x*
y*ne1*ne2*ne3*ne4*up)
This place receives and sends
the information of the next
step of the airplane process.
New stopover without
cleaning airplane=product(ac*sa*p*h
*te1*te2*te3*te4*nt*a*q*n*s)
This place receives a token
whether the airplane does not
have to be cleaned, which
means the airplane will do the
next flight without the need of
cleaning.
Aircraft in AOG airplane=product(ac*sa*p*h
*te1*te2*te3*te4*nt*a*q*n*s)
This place receives the token
whether the airplane breaks
down and needs major
reparation.
Counter ne=product (u*d*tr*cu*ci*ex)
This place keeps track of the
number of times the aircraft
has been cleaned.
Solution airplanes1=product(p*h*te1
*te2*te3*te4*nt*s*a*q*n*b*x*
y*ne1*ne2*ne3*ne4*up)
This place records the final
state of the aircraft.
Cleaning airplane=product(ac*sa*p*h
*te1*te2*te3*te4*nt*a*q*n*s)
This place records the
information necessary to
decide if airplane has to be
cleaned.
A = { A1, A2, … , Ana} represent the directed
arc set, which relate transition and place nodes
such as A P T TP
N = It is the node function N(Ai), which is
associated to the input and output arcs. If one is
a place node then the other must be a transition
node and vice versa.
C = is the colour set functions, C(Pi), which
specify for the combination of colours for each
place node such as C: P .
()
ij
CP C ,
ij
PPC
G = Guard function, it is associated to transition
nodes, G(Ti), G: T EXPR. It is normally
used to inhibit the event associated with the
transition upon the attribute values of the
processed entities.
E = these are the arc expressions E(Ai) such as
E: A EXPR. For the input arcs they specify
the quantity and type of entities that can be
selected among the ones present in the place
node in order to enable the transition. When it is
dealing with an output place, they specify the
values of the output tokens for the state
generated when transition fires.
I = Initialization function I(Pi), it allows the
value specification for the initial entities in the
place nodes at the beginning of the simulation.
It is the initial state of a particular scenario.
EXPR denotes logic expressions provided by
any inscription language (logic, functional, etc.)
The state of every CPN model is also called the
marking which is composed by the expressions
associated to each place p and they must be
closed expressions i.e. they cannot have any
free variables.
6.2. Model Definition
The model is divided into two main modules:
1) Decision-making: the necessary information is
collected to decide what the model is going to do. The
results of this decision are:
The plane does not have to be cleaned.
The plane has to be cleaned.
The plane has suffered a problem.
An extra cleaning has been requested.
2) As soon as the decision has been taken, the plane shall
be sent to the corresponding section of the model to
execute the next task. The variability is integrated in the
model through the use of two variables that simulate the
situations that the plane undergoes a breakdown or an
extra cleaning is requested.
The developed model in CPN is composed by 11 place
nodes and 5 transition nodes. Table 2 describes the place
nodes of the model.
Table 2: Place Nodes
Table 3 presents the colour definition used in the
CPN model of the new cleaning system.
Table 3: Colours and Definitions
Colour Definition
Ac Aircraft identification.
Sa Flight identification.
H Amount of minutes that the aircraft
has flown since the last cleaning
service.
P The total of passengers that has
been transported since the last
cleaning service.
te1 Whether the stopover is in the first
group of the table 1.
te2 Whether the stopover is in the
second group of the table 1.
te3 Whether the stopover is in the third
group of the table 1.
te4 Whether the stopover is in the
fourth group of the table 1.
Nt The type of cleaning that was done
last time.
A The amount of minutes that the
aircraft has flown sinc e the last
cleaning number 5.
Q Whether the plane can ca rry out all
types of cleaning.
Table 3 (cont.)
Colour Definition
N The amount of minutes that the
airplane has flow n since the last
cleaning number 4.
S The nu mber of flights has flown the
aircraft, since t he last cleaning
service.
Up The operational status of the
aircraft.
E Whether an extra cleaning has been
requested.
sa2 Next flight identification.
h1 The duration of the next flight.
p1 The number of passenger s will be
transported on the next flight.
ne1 Whether the next st opover is in the
first group of table 1.
ne2 Whether the stopover is in the
second group of table 1.
ne3 Whether the stopover is in the third
group of table 1.
ne4 Whether the stopover is in the
fourth group of table 1.
U Cleani ng counter of type 1
D Cleani ng counter of type 2
Tr Cleaning counter of type 3
Cu Cleaning counter of type 4
Ci Cl eaning counter of type 5
The model has been run using the information of a
particular day in which the airline had 27 operative
aircrafts.
Figure 1 presents transition T1, which would receive the
information related to the actual and future flights, the
operational status of the incoming aircraft and whether
an extra cleaning has been requested.
The outcome information of the transition will be used to
decide the next step of the airplane.
Figure1: Transition T1
Arc (1): This arc has the restrictions to decide the next
step of the airplane. It evaluates the operational status of
the aircraft, whether is necessary a cleaning service or it
has being requested an extra cleaning. The outcome
information will assign what the next step of the aircraft
is. This information is evaluated by the second transition.
Figure 2 illustrates transition T2; it receives the
information about what the next step of the aircraft will
be and based on that information it will send the aircraft
to the corresponding place.
Figure 2: Transition T2
Arc (2): this arc evaluates the restrictions related to what
type of cleaning will be performed in the airplane and it
will increase the value of the correspondent cleaning
counter. The place node contains the information about
which flight must be cleaned.
Arc (3): this arc send the current status of the data
information to the correspondent place node
(SOLUTION). The income data contains the information
of the flight that must be cleaned and the next flight. The
SOLUTION place node keeps track of the current status
of the system.
Arc (4): this arc evaluates the information of the tokens
concerning what type of cleaning operation shall be
performed. The decision takes into account the stopover
time, the information of the airplane and the information
of the flights. The outcome information will assign the
type of cleaning to be performed. The information is used
by the fourth transition.
The Figure 3 presents transition T3. This transition
represents the outcome when the airplane does not need
a cleaning service. The data will be updated with the
information of the next flight and passed through with the
token colours to the AIRPLANES place node.
Figure3: Stopover without cleaning
Figure 4 presents transition T4, it evaluates the variables
to assign a cleaning in the next stopover. The data will be
updated using the token created in the AIRPLANES
place node.
Figure 4: Stopover with cleaning
Finally Figure 5 presents transition T5. This transition
evaluates the correspondent variables and simulates an
AOG to the correspondent Aircraft. Once the AOG has
been performed, the variables’ data is updated through
the correspondent token created in the AIRPLANES
place node.
Figure 5: Solve AOG
6.3. Analysis of the causal model.
To evaluate the proposed system, the model was
simulated 15 times. Table 4 presents the results obtained
with the simulation.
Table 4: Simulated Results from the causal model
Results 1 2 3 4 5 6 7 8 9 1
0 1
1 1
2 1
3 1
4 1
5 Aver
age
value
Cleaning
1
1
7 1
8 1
8 1
8 1
8 1
8 1
8 1
8 1
8 1
8 1
9 1
8 1
7 1
7 1
8 17,93
Cleaning
2
1
5 1
4 1
4 1
4 1
5 1
4 1
5 1
5 1
5 1
5 1
6 1
5 1
5 1
5 1
5 14,80
Cleaning
3
1
1 1
1 1
0 1
1 1
1 1
2 1
1 9 1
1 1
1 1
1 1
1 1
1 1
1 1
1 10,87
Cleaning
4
3 3 4 3 2 3 3 3 3 3 3 3 3 3 3 3
Cleaning
5
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Extra
Cleaning
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
On the other hand, it is possible to evaluate the cost
impact of implementing the new schema. The cost
analysis can be appreciated in Table 5.
Table 5: Economic analysis
Economic
Variables
Actual System Proposed
System
Difference
between
Systems
Number of
cleaning
flights
1978 1310 668
Number os
extra cleanings
133 18 115
Percentage of
cleaned flights
43,84% 29,03% 14,81%
Cost of the
cleaning
service *
€ 49.421 € 24.675,8 € 24.745,20
Airport Rates € 37.895,99 € 34.102,69 € 3.793,30
Percentage of
delayed flights
19,32% 9,64% 9,68%
Number of
delay flights
493 50 443
The cost of
delay flights
€ 4.317,66 € 340,87 € 3.976,79
RESULTS € 91.634,65 € 59.119,36 € 32.515,29
Through the results, it can be concluded that the
proposed model is less expensive than the actual model
due to:
Creating more types of cleaning with different
durations makes the cleaning service more
flexible which means the cleaning service has
been adapted to the stopovers time. The number
of cleaning operations has been reduced due to
the flexibility achieved. With the proposed
model only the 29,03% of flights were cleaned
rather than 43,84% of the current schema.
The amount of delay flights has been reduced.
With the proposed model only the 9,64% of the
flights were delayed by the cleaning service
rather than the 19,32% of the current schema.
The proposed schema decreases the number of
extra cleanings. With the proposed model it is
needed 18 extra cleaning rather than 133 extra
cleaning in the current system.
The total amount of cost in the proposed model
for November's month is € 59.119,36 rather the
€ 91.634,65 of the current schema.
7. VALIDATION OF THE MODEL
The previous results have been validated using a
discrete-event-oriented simulation program (SIMIO) in
which the complete elements of the Turnaround of a
A320 aircraft have been taken into account. The purpose
of the simulation model is twofold, on the one hand to
evaluate the results of the CPN causal model and on the
other to include all the elements of an actual turnaround
that could not be included in the causal model. The final
goal is to obtain a better management for the turnaround
process that allows mitigating the delays caused by the
current management schema.
Figure 6 shows a snap shot of the graphical aspect of the
model for the turnaround of the Aircrafts of the company
(Airbus-A320).
Figure 6: Virtual environment
Several operations occur during the stop over: Catering,
Fuelling, Disembarking-Boarding of Passengers,
Cleaning. In Figure 6 three trucks can be appreciated, 1
big truck performs the catering operation, the one under
the wing is fuelling the aircraft and the one in the rear
position of the aircraft is cleaning the system from
organic disposals. It can also be appreciated that the
passengers are deboarding the plane through the fingers.
Is important to note that in the particular case of the
fueling operation it does not start until all the passengers
have left the aircraft; this is due to security reasons.
In the turnaround process some activities has been
identified as being the critical path of the turnaround
time. Figure 7 illustrates the total operations that can be
performed in such an aircraft and the ones that are part of
the critical path of the process (AIRBUS 2012).
Figure 7: Operations of the turnaround for a A320
The airline of the study does not perform all the
operations; in order to reduce the turnaround time the
company perform only a few operations, namely
boarding, deboarding, catering on door R2 only,
cleaning, refuelling, cargo operations and toilet
servicing. Under this operative schema the cleaning
operation becomes part of the critical path that
determines the turnaround time of the aircrafts.
Parameters of the Simulation Model
In order to assess the importance of the cleaning
operations, the current process was simulated using
information provided by the airline. Table 6 presents the
values used for the model.
Table 6: Simulation parameters
Operation Time
Opening/closin g doors 2 min
Deboarding Rate 22 pax/min
Deboarding Rate/pax Triangular (2.5,2. 7,3) sec
Boarding Rate 18 pax/min
Boarding Rate/pax Triangular (3,3.3,5) sec
Fueling Time Triangular (7,8, 9) mins
Cleaning Operation Triangular (8,13,16) mins
Full size trolley equivalent (FSTE)
to unload/load
7 for R2
Load Time of each Trolley 1.5 min/FSTE
Catering Equipment
Position/Removal
2 min
Probability of Cleaning 0.4348
Probability of Extra Cleaning/P.of
Cleaning
0.0664
The previous data was used for developing the
turnaround model for the current and the proposed
schema. The last two rows were obtained from the
information provided by the causal model. The first value
(P. of Cleaning) is the probability that the aircraft
performs a cleaning operation; and the second value
corresponds to the conditional probability of an extra
cleaning once the cleaning has been performed. The rest
of the values will be the same for the current scenario and
the proposed one.
7.1. Evaluation of the Proposed Schema
The simulation model was used for analysing the
current operations and at the same time obtaining
different values that provide insight about the
inefficiencies present in the current performance. The
second scenario will be implemented assuming new
values for the cleaning operations (based on the results
provided by the causal model).
Current Operations
The simulation model was run with the aforementioned
values and the turnaround times, number of
extracleanings and delays were analyzed. Table 7
presents the results obtained with the current operations.
Table 7: Information from the current process
Cleaning Operation
AVG Min. Max. STD. Dev.
Max. No. of
Extra Cleanings
6.7 3 12 2.306
Max.No. of
Total Delays
37.43 13 81 15.904
Turnaround
Times
38.59 37.17 40.9 0.8262
Max.
Turnaround
Times
54.38 49.14 59.31 1.8539
The previous values were obtained of a total of 240
flights and the simulator was run for 30 replications. As
it can be appreciated the first row gives information
about the maximal number of extra cleanings, the second
row about the maximum number of total delays and the
last two rows the average and maximal turnaround times
for this scenario.
In the case of the delays it should be pointed out that the
upper bound of delays is 81 out of 240 flights which
correspond approximately to 33% of the scheduled
flights incurred in a delay. On the other hand the maximal
turnaround times which are the upper bound for the
model mean that some aircrafts could have a turnaround
time of 59 minutes which would be translated into a big
cost penalty for the airline.
Proposed Schema
The new schema was tested using the same values of the
standard operations but in this case the probability of the
cleaning operation and the conditional probability of an
extra-cleaning once the cleaning operation has been
performed are 0.2903 and 0.0137 respectively.
The cleaning times in this new schema also change to a
Triangular(5,7,8) since it is assumed that the aircraft in
the simulation model are only of group type I. Table 8
presents the results obtained with the proposed schema.
Table 8: Proposed Scenario
Cleaning Operation
AVG Min. Max. STD. Dev.
Max. No. of
Extra Cleanings
1.65 1 4 0.8846
Max.No. of
Total Delays
12.42 1 56 18.69
Turnaround
Times
37.57 36.03 39.47 0.9127
Max.
Turnaround
Times
40.4 38.57 43.47 1.235
From the previous table it can be appreciated that the
mean average turnaround time has been reduced about a
minute. As it will be clear with the next figure, the most
important achievement is that the dispersion or
variability is drastically reduced. As a consequence the
probability of delays has been reduced as it can be
appreciated in Figure 8.
Figure 8: The reduction in the avg. turnaround times
With the new implementations and with the dispersion
obtained from the simulation model, it can be appreciated
that the curve of the new schema falls within the
acceptable region while with the current operations
approximately 33% of the flights incur in delays.
On the other hand, if the worst-case scenarios are
analysed (i.e. the max. turnaround times) the
improvements are more evident. As it can be appreciated
from Figure 9, the worst-case values from the current
operations fall out of the accepted region while with the
new schema only approximately the 50% of the worst-
case turnaround times would incur in a delay.
Figure 9: The worst-case scenarios
8. CONCLUSIONS
In this article a new cleaning schema for an airline
was devised with the objective of reducing the costs of
extra-cleanings and to avoid as much as possible the
probability of delays in the turnaround time. The
proposed schema has been analysed using a causal model
developed using the coloured Petri net formalism and it
has been validated with a more detailed simulation model
that takes into account all the different operations that are
critical for the turnaround time. The results clearly
indicate that it is possible not only to reduce the extra-
cleanings which is a common practice for a commercial
airline but also reducing the possibility of incurring in
delays due to the cleaning operations.
ACKNOWLEDGMENTS
The authors would like to thank the Mexican
Council of Science and Technology (CONACYT) for
supporting the research, and UNAM-DGAPA project
PAPIIT IN116012-3.
APPENDIX A
Definitions
Chocks: A block or wedge placed under the aircraft
wheels, to keep it from moving
Medium haul flights: is a flight between 3 and 6 hours in
length.
On board cleaning service: is the main job in cabin
service. They include task such as cleaning the passenger
cabin, replenishment of on-board consumables or
washable items such as soap, pillows, tissues and
blankets, and do the sanitation service.
Short haul flights: is flight: is a flight under 3 hours in
length.
APPENDIX B
Operational Costs
Airport Rates Every 15
minutes
Every 30
mins
Per
Flight
Monthly
Airport Use € 9,88
Vehicle Parking € 0,02
Workers 33,90
0.00
0.10
0.20
0.30
0.40
0.50
30 32 34 36 38 40 42 44
PROBABILITY
MINUTES
Avg.TurnaroundTimes
CurrentOperation
ProposedOperation
DELAYED
TURNAROUND
0.00
0.10
0.20
0.30
0.40
0.50
30 35 40 45 50 55 60 65
PROBABILITY
MINUTES
MaximalTurna rou nd Times
CurrentOperation
ProposedOperation
DELAYED
TURNAROUND
Energy
system 400HZ
€ 6,79
Fingers use € 27,18
rate for cleaning 1, 2
and stopover cleaning
€ 9,90
rate for cleaning 3 and
Overnight cleaning
€ 43,87
rate for cleaning 4 € 77,86
rate for cleaning 5 and
3 and deep cleaning
€ 111,82
Delays € 2,27
REFERENCES
AENA air tariffs. Available from
http://www.aena.es/csee/ccurl/124/479/guiaTarifasNA_
2013-EN.pdf. [accesed May 20, 2013]
Airbus, “A320, 2012. Aircraft Characteristics Airport
and Maintenance Planning”,Technical Report,
Airbus.
Bazargan, M., 2004. Airline operations and Scheduling.
Burlington, USA, Ashgate publishing company.
Christensen, S., Jensen, K., Mailund, T., Kristensen,
L.M., 2001. State Space Methods for Timed
Coloured Petri Nets. Proc. of 2nd International
Colloquium on Petri Net Technologies for
Modelling Communication Based Systems, 33-42,
Berlin.
Civil Aviation Department Hong Kong, China, 2012. CAD
452 Aircraft Maintenance Schedules and
Programmes, Information and Guidance. Available
from
http://www.cad.gov.hk/english/pdf/CAD452.pdf uk
[accessed June 15, 2013].
CPNTools Available from http://www.cpntools.org.
Guest, T. 2007. Air traffic delay in Europe. Trends in Air
Traffic Vol. 2, Brussels-Belgium,
EUROCONTROL.
Jensen, K., 1997. Coloured Petri Nets: Basic Concepts,
Analysis Methods and Practical Use. Springer-
Verlag, Berlin.
Jetsky, M, 2009. The propagation of air transport delays
in Europe, Thesis, RWTH Aachen University.
Moore, K.E., Gupta, S.M., 1996. Petri Net Models of
Flexible and Automated Manufacturing Systems: A
Survey. International Journal of Production
Research, 34(11), 3001-3035.
Rhodes, W., Lounsbury, R., Steele, K., Ladha, N., 2003.
Fatigue Risk Assessment of Aircraft Maintenance
Tasks. Transportation Development Centre Safety
and Security Transport Canada.
Yeung, S.S.M, Yu, I. T. S., Hui, K.Y.L., 2005. World at
Work:Aircraft cabin cleaning. Occupational and
Environmental Medicine, 62:58-60.
AUTHORS BIOGRAPHY
Miguel Mujica Mota was born in Mexico City. He
Studied Chemical Engineering in the Metropolitan
Autonomous University of Mexico. He also studied a
MSc. in Operations Research at the National
Autonomous University of Mexico. After spending some
years in industry he continued his studies and he obtained
the PhD in Industrial Informatics in 2011 with the
highest honors from the Autonomous University of
Barcelona and the PhD in Operations Research at the
National Autonomous University of Mexico.
Dr. Mujica has been awarded with the Candidate
to Level I of the Mexican Council of Science and
Technology where he also participates as a scientific
evaluator. He is currently the sub director of the
Aeronautical Studies at the Autonomous University of
Barcelona and his research interests lie in the use of
simulation, modeling formalisms and heuristics for the
analysis of performance and optimization in
manufacture, logistics and aeronautical operations.
Mireia Soler Grané was born in Barcelona, Spain.
She studied Aeronautical Management in Air Transport
Logistics. While she was studying, she was working in
Barcelona’s airport as Handling Agent, after two years
she was the Assistant of Station Manager of Spanair S.A.
Idalia Flores received a Master with honors, being
awarded the Gabino Barreda Medal for the best average
of her generation, in the Faculty of Engineering of the
UNAM, where she also obtained her Ph.D. in Operations
Research. Dr. Flores is a referee and a member of various
Academic Committees at CONACYT as well as being a
referee for journals such as Journal of Applied Research
and Technology, the Center of Applied Sciences and
Technological Development, UNAM and the
Transactions of the Society for Modeling and Simulation
International. She is a full time professor at the
Posgraduate Program at UNAM and her research
interests lie in simulation and optimization of production
and service systems.
... This study has been performed for a LCC in Spain (Mujica et al. 2013) in which a new cleaning schema was proposed for reducing the turnaround time. We identified that with different types of cleaning services based on the flight history of the aircraft, it was possible to achieve important reductions in the turnaround time. ...
Conference Paper
Full-text available
the aviation industry is a changing industry in which several factors influence the performance of the airport and the network of airports that are interconnected. Business models, technical operations in airspace and in the airfield, societal conditions among others are some of the ones that must be taken into account in order to get a fully understanding of the cause-effect relationships that hinder the proper management of the system. In recent times with the evolution of the computer technology and the level of maturity of the algorithms used to simulate and analyze dynamic systems, simulation has gained more importance than before. Simulation approaches emerge as the ones that are able to take into account the stochastic nature of dynamic systems besides all the different factors that impact the systems under study. This is something that traditional analytical approaches could not evaluate and therefore under the constant change of the systems they lack of the proper flexibility to provide timely solutions. However with the popularity that simulation has gained, the different steps and good practices that must be taken into account are commonly forgotten when the simulation model is developed and then the system is analyzed. One of the key characteristics of simulation is that it is able to integrate the variability and uncertainty inherent in all these systems. This variability can lead to different conclusions when the systems are assessed and it is a common flaw to ignore the consequences of in addition to characterizing the performance of systems on individual and punctual values such as average values for performance indicators. The current paper addresses some of the common flaws or pitfalls incurred when simulation is used for analysis of aeronautical systems. The flaws are exemplified through different cases in which the conclusion from the analysis might differ depending on the angle of the analysis performed thus with different economic consequences for the decision makers. The paper illustrates some of the flaws with different studies performed by the author in which the afore-mentioned problems could have been incurred if not properly performed the simulation methodology. It is expected that this paper serves as an eye-opener for avoiding these common pitfalls in industrial applications.
Article
Full-text available
The aviation industry is a changing industry in which several factors influence the performance of the airport and the network of airports that are interconnected. Business models, technical operations in airspace and in the airfield, societal conditions among others are some of the ones that must be taken into account in order to get a full understanding of the cause-effect relationships that hinder the proper management of the system. In recent times with the evolution of the computer technology and the level of maturity of the algorithms used to simulate and analyse dynamic systems, simulation has gained more importance than before. Simulation approaches emerge as the ones that are able to take into account the stochastic nature of dynamic systems besides all the different factors that impact the systems under study. This is something that traditional analytical approaches could not evaluate and therefore under the constant change of the systems, they lack the proper flexibility to provide timely solutions. However with the popularity, that simulation has gained, the different steps and good practices that must be taken into account are commonly forgotten when the simulation model is developed and then the system is analyzed. The current paper addresses some of the common flaws and pitfalls incurred when simulation is used for analysis of aeronautical systems. Pitfalls classification and suggestions for avoiding them are presented. Some flaws are exemplified through cases in which the conclusion from the analysis might differ depending on the angle of the analysis performed with the implications of different economic consequences for the decision makers. The paper also illustrates some of the flaws with two examples performed by the authors in which the aforementioned problems could have been incurred if the simulation methodology had not been properly performed. It is expected that this paper serves as an eye-opener for avoiding these common pitfalls and flaws in aviation practice.
Article
Full-text available
A look from the inside Most of us have the experience of travelling on aeroplanes and are impressed by the cleanliness and orderliness of the cabin that welcome us together with the friendly and smiling faces of the aircrew every time we board an aeroplane. The cleanliness and orderliness of the cabin are the results of the hard work of heroes behind the scene—the aircraft cabin cleaners. The work of these cleaners seldom attracts our attention, and in fact, very few of us have directly met with them before. Aircraft cabin cleaning is a unique type of cleaning work characterised by a high concentration of physical activities in time and space that are not fully under the direct control of the service providers and their workers. The confined workspace (due to maximisation of carrying capacities and comfort of passengers) and the tight work schedule (especially with short stay flights) are potential underlying risk factors for the development of musculoskeletal disorders among the cleaners. In this article, we describe the occupational health aspects of aircraft cabin cleaning at Hong Kong International Airport. Three major companies provide the cabin cleaning services for all aircraft in the airport, and the total number of workers employed is around 1200. Airline companies usually contract out cabin cleaning to aircraft services companies. There are basically three levels of cleaning services for the aircraft: quick transit cleaning, overnight (lay-over) cleaning, and deep cleaning. The types of services rendered depend on the duration of stay of the aircraft at the airport. The tasks in a typical quick transit cleaning involve: (1) seat cleaning, (2) seat pocket cleaning, (3) ashtray cleaning, (4) galley cleaning, (5) toilet cleaning and replenishment, (6) floor cleaning, and (7) blanket management. Lay-over cleaning is more thorough and in addition to the above processes includes: …
Article
This book demystifies operations and scheduling within airlines, presenting easy-to-understand models, applied to practical examples. After introducing the key issues, it offers an objective review of the various optimization models adopted in practice. Each model provides solutions to a range of scenarios, with case studies similar to those experienced by commercial airlines. Using unique source material and interviews with alumni working at operations and scheduling departments of various airlines, this solution-orientated approach has been used on many courses with outstanding feedback.
Article
Petri nets (PNs) have recently emerged as a promising approach for modelling flexible and automated manufacturing systems. PNs are a graphical and mathematical modelling technique that is useful for modelling concurrent, asynchronous, distributed, parallel, nondeterministic, and stochastic systems, as attested by a steady stream of papers which appeared throughout the 1980s. Since 1989, there has been an explosion of interest in using PNs to model, simulate, and analyse manufacturing systems. We present an overview of PN theory. We then present a schema for categorizing PN models of manufacturing systems, followed by a survey of PN models of flow lines, automatic transfer lines, job shops, flexible manufacturing systems, and assembly systems. This discussion represents the most comprehensive survey of applications of PNs to manufacturing through 1994. Finally, we present a summary and some conclusions.
Article
We present two recently developed state space methods for timed coloured Petri nets which reconciles state space methods and time concepts of coloured Petri nets. The first method is based on an equivalence relation on states which makes it possible to condense the usually infinite state space of a timed coloured Petri net into a finite condensed state space without serious loss of analysis power. The second method, the sweep-line method, exploits the increasing model time of timed CPnets to store only small fragments of the full state space in memory at a time thereby alleviating the the state explosion problem. The performance and potential of the methods is demonstrated on an industrial case study by means of a computer tool implementing the two methods.
The propagation of air transport delays in Europe, Thesis
  • M Jetsky
Jetsky, M, 2009. The propagation of air transport delays in Europe, Thesis, RWTH Aachen University.
Air traffic delay in Europe
CPNTools Available from http://www.cpntools.org. Guest, T. 2007. Air traffic delay in Europe. Trends in Air Traffic Vol. 2, Brussels-Belgium, EUROCONTROL.
A320, 2012. Aircraft Characteristics Airport and Maintenance Planning
  • Airbus
Airbus, "A320, 2012. Aircraft Characteristics Airport and Maintenance Planning",Technical Report, Airbus.
Fatigue Risk Assessment of Aircraft Maintenance Tasks. Transportation Development Centre Safety and Security Transport Canada
  • W Rhodes
  • R Lounsbury
  • K Steele
  • N Ladha
Rhodes, W., Lounsbury, R., Steele, K., Ladha, N., 2003. Fatigue Risk Assessment of Aircraft Maintenance Tasks. Transportation Development Centre Safety and Security Transport Canada.