MODELING, SIMULATION AND OPTIMIZATION OF THE MAIN PACKAGING LINE

OF A BREWING COMPANY

N. Basán (a), L. Ramos (b), M. Coccola (c), C. A. Méndez (d)*

(a) (b) (c) (d)

INTEC (UNL - CONICET), Güemes 3450, 3000 Santa Fe, Argentina.
(a)
[email protected], (b)[email protected], (c)[email protected], (d)*[email protected]

ABSTRACT order to improve the global efficiency. In this way, a

Discrete event simulation (DES) techniques cover a comprehensive simulation-based model was developed
broad collection of methods and applications that allow so that different future commitments and changing
imitating, assessing and predicting the behavior of market conditions can be easily suited in the medium or
complex real-world systems. The main purpose of this short term.
work is to develop a novel DES model to optimize the
design and operation of a complex beer packaging 2. METHODOLOGY
system in order to perform a sensitivity analysis to find Having stated the general and specific objectives of the
one or more alternatives to increase productivity levels. project, all the necessary data from the company under
In this way, advanced technologies of modeling, studying was collected by using five different
simulation and optimization for system design and techniques: (i) staff interviews, (ii) in-situ observation,
operation are applied. The model is developed by using (iii) historical data collection, (iv) reading manuals, and
the DES tools provided by the SIMIO simulation (v) review of equipment and transport.
software. The proposed tool is able to carry out Once the required information was collected, this
evaluations of the system using a 3D user-friendly one was analyzed, filtered and documented. Such
graphical interface that shows the dynamic evolution of procedure allows us to identify critical points and
the system over time. By using the proposed simulation potential problems to be solved in the current and
model, the results of this paper illustrate how the levels desired situation of the packaging process.
of productivity may vary by reducing micro-downtime After understanding the real system that is subject
of machines, when transport rates and other problem of our simulation study, the conceptual model was
features are properly changed with them. developed. If we need to develop a model on a
simulator, we need to determine the level of abstraction
Keywords: simulation, optimization, packaging line, at which to work. This process of abstracting a model
brewing company from the real world is known as conceptual modeling.
For this particular project, SIMIO simulation software
1. INTRODUCTION was chosen because this one allows to build animated
In modern production processes, quality takes models in three dimensions (3D), facilitating the
precedence in relation to production volume. A verification and validation of the simulation model.
fundamental topic in market requirements is the product The inherent advantages of the simulation model
presentation. It implies that production lines need an developed were highlighted by solving three scenarios:
additional number of machines to perform a wider (i) theoretical or ideal, (ii) current, and (iii) suggested
variety of tasks. In this way, all activities must be scenario. A sensitive analysis had to be conducted to
conducted with the highest possible quality. Therefore, determine the more suitable alternatives regarding to the
the growing demand and specialization in the overall performance of the company, and consequently,
presentation of the product make packaging lines more the expected economical benefits.
complex. More diversified tasks are performed on them.
This work arises from the need to identify, analyze 2.1. Production Process
and reduce the causes affecting the productivity of the The beer production process comprises a series of
main packaging line of an international beer company manufacturing steps depending on the type of beer,
located in the province of Santa Fe, Argentina. varying the amount and type of raw material. Such steps
Currently, the efficiency of the company’s lines is lower are: malting, malt milling, mashing, cooking, wort
than the level suggested by the managers. This situation cooling and clarification, fermentation, maturation, and
has a directly impact on the current production level due the packaging process at the end.
to the packaging process is an essential step in the
whole production process. 2.1.1. Packaging Process
This work aims to optimize the design and A packaging line involves a set of machines, equipment
operation of the main packaging line of the company in units and tools needed to perform the process

Proceedings of the European Modeling and Simulation Symposium, 2013 551

978-88-97999-22-5; Bruzzone, Jimenez, Longo, Merkuryev Eds.
 operations. The success of the line depends on proper unpacking. The machine is designed for gripping and
coordination of different elements. Similarly, in the case moving sets of bottles into crates synchronously
of a packaging line of a brewing company, beer and entering the computer.
containers move through a series of processing stages Finally, we proceed to make palletizing, i.e.
until the final product is obtained. The flowchart is drawers are placed on a wooden stand known as pallet
shown in Figure 1. or pallet for easy handling and transportation future.
The arrangement of the crates on the pallet is performed
in layers according to a set distribution, in order to form
a compact load unit capable of supporting stable after
storage, transport and distribution.
On the other hand, different units are involved in
the packaging process using transport. These are
between the various machines in the line, providing a
connecting element and synchronism between two of
Figure 1: Beer packaging process them. The rate is fixed by the same variable speed
drives, and startup and shutdown is done by proximity
The first operation performed in a beer packaging sensor and optical detectors. The elements considered
line is depalletizing and disassemblying the pallet that along the packaging process are pallets, crates and
comes from the deposit with empty bottles. bottles.
The next step, 100% control, aims to eliminate
most of the waste entering with the drawers. This 2.2. Simulation Model
inspection is carried out manually by an operator. He Process simulation is one of the most useful tools of
removes items or bottles that could possibly harm the industrial engineering, which is used to represent a
following machines. complex process by another which makes it simple and
Then, the process proceeds to perform the understandable.
unpacking by a computer that has the function of The proposed simulation model was developed by
extracting the empty bottles boxes that feeds the using the SIMIO programming package, which is one of
packaging line. The objective of the operation is to the most specialized software in the area of process
separate the drawer containers for subsequent washing simulation that minimizes the risk and uncertainty in
operations on the respective machines. decision making, as well as minimizing the costs by
After unpacking, the bottles are guided through the improving the use of resources, reduced time spent and
transport system to the washing process. Because all the minimization of the probability of risk.
returnable bottles should be sanitized before being filled Likewise, the packaging process involves five
with beer again, the goal of this stage is to perform the classes of modeling elements:
physical and biological cleaning, removing all dirt,  Bottles that run the line
labels, adhesive and foil.  Crates of bottles
While we can assume that all bottles are dirt free in  Machines that perform the operations
the washing machine, there is a risk this stage has not necessary to prepare and process the product
been able to completely remove all cleaning agents.  Transports located between different machines
Therefore, the output of this machine is a containers that make up the line
inspector.  Operators who have assigned different tasks.
Next, we proceed to perform filling and topped
with beer in containers. The aim of the operation is to 2.3. SIMIO Simulation Software
transfer product from a pot bulk and individual SIMIO is modern flow simulation software for discrete
containers with airtight lids seal rolled steel to ensure event processes, and procedures based on objects.
durability, quality and inviolability of beer. SIMIO allows conducting a simulation project in a
Subsequently, through the process of pasteurization is much shorter time than usual. It is the first software that
achieved that beer is kept in ideal state at least until the combines simulation modeling speed allowed by the
date of minimum durability, i.e. the primary objective object-oriented technology with the flexibility and
of pasteurization is to avoid possible biological power of procedures. This software enables the modeler
decomposition and lengthen the bottled product. to build animated models in three dimensions (3D) in
The next step aims to place the labels presented in one third of the usual time, and thus frees up time to
the final product. The labeling process begins when devote to analysis of alternatives and scientifically
filled and capped bottles entering the labeling machine, informed decision-making.
and ends with a level-cap inspection rejecting bottles Therefore, to model the packing process, it is
that do not meet any of the required characteristics in necessary to represent the major components of the
terms of filling level, internal pressure, and missing system, i.e. the products (bottles, boxes, pallets),
state missing labels and cap. machines, inspectors, transport and accumulation tables
Once the bottle labeling operation is performed that form.
encased, which is contrary to the operation of

Proceedings of the European Modeling and Simulation Symposium, 2013 552

978-88-97999-22-5; Bruzzone, Jimenez, Longo, Merkuryev Eds.
 Different views of the simulation model of the The “Source” module is used to create entities that
main packaging line, developed in the SIMIO arrive to the system. Figure 6 shows as pallet entity
simulation software, are shown in Figures 2-4. arrival is defined.

 
Figure 6: Pallet entity arrivals.
Figure 2: 3D SIMIO model (despalletizer)
2.4.2. Depalletizer and Unpacker Machines
The depalletizer is the first equipment unit in the
packing process. The module “Separator” provided by
SIMIO is used to represent the depalletizer’s operation,
which is shown in Figure 7. On the right side of this
picture, we can see the properties associated with the
“Separate” module, i.e. the property “Processing Time”
determines that each pallet is processed in a time of 50
seconds. Each pallet that is full of drawers enters to
depalletizer to be processed and then 50 new entities
representing the drawers are generated by the model.
Figure 3: 3D SIMIO model (unpacker)

Figure 7: 2D SIMIO model (depalletizer machine)

Figure 4: 3D SIMIO model (palletizer)
The unpacker machine is modeled in the same way
2.4. Creation of Simulation Model that depalletized equipment. Each drawer that is full of
For the creation of the simulation model, standard bottles enters to unpacker to be processed and then 12
elements as source, process and sink were used, new entities representing the bottles are generated by
connected by a set of paths. The following components the model. Figure 8 shows the “Separator” module
were identified in the real packaging line: associated to the operation of unpacker unit.
2.4.1. Pallets, Boxes and Bottles
As shown Figure 5, the dynamic entities moving
through the system are: (i) pallets, (ii) drawers, and (iii)
bottles.

Figure 8: 2D SIMIO model (unpacker machine)

Between two operations describe above, there is an

Figure 5: System entities
Inspection Process (Control 100%) that controls the

Proceedings of the European Modeling and Simulation Symposium, 2013 553

978-88-97999-22-5; Bruzzone, Jimenez, Longo, Merkuryev Eds.
 drawers entering to the line. The representation in amount of filling valves). The capping machine, which
SIMIO of this process is shown in Figure 9. As we can is then, has the same processing capacity too.
see in the picture, a “Decide Step” that uses a
probabilistic distribution is defined so that defective
entities can be rejected.

Figure 11: 2D SIMIO model (bottles inspector)

Figure 9: 2D SIMIO model (Control 100%)

2.4.3. Bottles Washer

This operation is built by placing 40 “Conveyor”
objects in the SIMIO model. Each conveyor represents
a real conveyor belt, which has a capacity of
transporting until 710 bottles and a fix speed assuring
that the bottles will be in the machine the minimum
required time (45 minutes). Figure 10 shows the
input/output logic of this stage. Figure 12: 2D SIMIO model (filling machine)

2.4.6. Pasteurizer Machine

The pasteurizing machine has two floors which were
represented in SIMIO by 60 conveyors working in
parallel (processed capacity of the equipment). In this
stage, the bottles cross by “rainfall areas” that give
water at different temperatures.

Figure 10: 2D SIMIO model (washing machine)

For the creation of the input logic to the washing

machine, standard elements as events and timer were
used. A group of 40 bottles enters to the washing
process every 2 seconds
Figure 13: 2D SIMIO model (pasteurizing machine)
2.4.4. Empty Bottles Inspector
This stage aims to verify the bottles that previously
have been processing in the washing machine. As
shown in Figure 11, a basic node is used to represent
this operation. Such node has one input path and three
outputs path. The first output path receives the bottles
that have a physical defect. The bottles that have some
dirt are sent by the second output path. Finally, the
accepted bottles continue their normal processing by the
third output path.

Figure 14: 2D SIMIO model (pasteurizing machine)

2.4.5. Filling Machine
This processing stage is represented by a conveyor that
has a transportation capacity of 154 bottles (equal to the

Proceedings of the European Modeling and Simulation Symposium, 2013 554

978-88-97999-22-5; Bruzzone, Jimenez, Longo, Merkuryev Eds.
 2.4.7. Labeler Machine tables assured that if an equipment is broken, the rest of
This equipment unit has an operation similar to the filler machines that are upstream can follow working.
so that both processes were modeled in the same way On the packaging line there are three accumulation
(see Figure 15). Two inspectors look at the bottles to the tables, two for bottles and one for drawers. The first
end of this stage. This control process is defined accumulation table is located between the empty bottles
similarly to Control 100% process described above. inspector and the filling machine (see Figure 18). The
Besides, in order to compute the total amount of second one is located between the pasteurizer
rejected bottles, two “Sink” component were used in the equipment and labeling machine. Finally, the drawer
simulation model (HUEFT and FT_50). accumulator is between the unpacker machine and the
packer machine.

Figure 15: 2D SIMIO model (labeler machine)

Figure 18: 2D SIMIO model (accumulation table for
2.4.8. Packer and Palletizer Machine bottles)
A “Combiner” module was defined in the simulation
model to represent the behavior of the packer and Figure 19 shows as a ”Monitor” element can be
palletizer machine (Figure 16 and Figure 17). In the first used to control the capacity of the conveyor that is
process, 12 bottles are assembled into a drawer. After above the accumulation tables for bottles. If the
that, the palletizer process puts together 50 drawers in a capacity of conveyor changes, a process called
pallet (10 drawers per stack, 5 stacks per pallet). Then, “Activar_Mesa” is trigger by the monitor.
the complete pallets are sent to storage modeled with a
“Sink” module.

Figure 19: 2D SIMIO model (Monitor element)

In addition, a binary variable named

Figure 16: 2D SIMIO model (packer process) “Activa_Mesa1” determines the current state of table 1.
If the table is working, active_mesa1 is equal to 1;
otherwise, it is set to zero.
Added to the above, the transport states located
before or after of buffer are monitored. In this way,
when table tapes are empty, the accumulating table is
disabled by stopping transports and assigning a value of
0 to the associated binary variable. Figure 20 shows the
monitor of one of the transports mentioned and the
process associated with the deactivation of the buffer.

Figure 17: 2D SIMIO model (palletizer process)

2.4.9. Accumulation Tables

The accumulation tables ensure a constant supply of
bottles or crates in the equipment that are after them.
Figure 20: Logic associated with deactivating the first
Since the machines are exposed to internal faults, these
accumulation table

Proceedings of the European Modeling and Simulation Symposium, 2013 555

978-88-97999-22-5; Bruzzone, Jimenez, Longo, Merkuryev Eds.
 2.4.10. Drawers Combiners
On drawers transport line there are two combiners
aiming to join two drawers belts into a single or
reversely. The first combiner is located after
depalletizer machine, more precisely where Control
100% is performed. Figure 21 shows as in this stage the
two drawers lines are combiner into a single belt. On
the other hand, the second combiner is situated before
palletizer machine and its function is to divide the
conveyor belt from the packer machine in two rows.

Figure 23: SIMIO simulation software (conveyor

properties)

On drawers line there are only single conveyors.

Instead, the bottle conveying line has conveyors of
different widths. Thus, from one to ten bottles can be
transported in parallel. An overview of this variable
capacity transport is given in Figure 24.

Figure 21: 2D SIMIO model (Drawers combiner)

Each combiner has a predefined logic, which is

defined from processes and is associated with the
transport involved. The first combiner transport has a
longer length and other shorter length in parallel. So it
allows passing more drawers with greater capacity in
order to achieve a balance in the accumulation of the
conveyors involved. It is worth to note that when one of Figure 24: SIMIO simulation software (bottle
them is moving, the other stops running. conveying line)

In order to join transports with different carry

capacities, several processes, whose logic is embedded
within “Basic Nodes” elements, were defined in the
simulation model (see Figure 25). Each process uses a
discrete probability so that the bottles can be distributed
on conveyors having available capacity. If any of the
selected conveyors is on the limit of its capacity, other
one in parallel must be chosen.

Figure 22: SIMIO processes

2.4.11. Transports
There are two transport lines, one for bottles and other
for drawers. “Basic Node” and “Conveyor” elements
were used in the simulation model to represent the two
transport lines. It is worth to remark that some
components of SIMIO have important parameters that
must be set by the user. In particular, some “Conveyor”
properties are given in Figure 23. From the picture, it Figure 25: 2D SIMIO model (distribution processes)
follows that these properties might be used to vary
things like conveyor speeds, traveler capacity, or the
option for accumulating or non-accumulating conveyors

Proceedings of the European Modeling and Simulation Symposium, 2013 556

978-88-97999-22-5; Bruzzone, Jimenez, Longo, Merkuryev Eds.
 2.5. Sensitivity Analysis
Therefore, after having identified the major operational
problems, a series of changes is proposed to the design
and operational model which a priori could increase the
production capacity of the company, increasing the
level of efficiency and address weaknesses the process.
This alternative scenarios raised by changing the values
of the factors considered most relevant.

Figure 26: SIMIO processes (Bottles distribution)  Scenario 1: system with engine capacities and
speeds and actual transport, and percentages of
2.4.12. Sensors rejections of inspectors,
Several sensors control the number of bottles or drawers  Scenario 2: system based on theoretical speeds
that are on the line. Such devices, located on strategic of the machines, provided the design of the
points of conveyor belts, emit signals so that transports line ("V Line").
or machines can start or stop their activities. These  Scenario 3: system considered in scenario 2,
sensors are switches that are activated or deactivated where, in addition, use is made of a battery
according to whether they are in contact with the object. drawer between the packer and unpacker
To represent the above behavior, “monitor” and machine,
“variable” elements were used in the simulation model  Scenario 4: system considered in scenario 2
so that the logic of each machine can be properly with the modification logic combiner boxes
defined. For example, the unpacker machine has three found after the depalletizing machine,
possible states: (i) stopped, (ii) low speed or (iii) high
 Scenario 5: system considered in scenario 2
speed. A variable was defined to determine the machine
with the increased transport speed that are in
state at a given time. The possible values of this
the area of the clean room due to the high rate
variable are: 0 (if the machine is stopped), 1 (if the
of accumulation of bottles with those found
equipment is operated at low speed) or 2 (if the machine
before the empty bottle inspector .
is running at high speed). In addition, three monitors
were defined for associated transports. If a capacity
Throughout the model, we adopt a series of
change is detected in them, the monitors trigger a
measures for evaluating performance goals achieved by
process determining the speed at which the equipment
the line and compare its performance against changes
should operate. This value is then saved in a predefined
that may occur. The proposed changes were modeled
variable. On one hand, if there is no accumulation in
using the tools Ape presents the simulator and that were
output transport and there are drawers in input transport,
mentioned earlier. Thus, modifications were made on
the machine operates at low speed. On the other hand, if
these variables, as they are directly related to the
there is accumulation in the input conveyor, the
operation of the packaging line and determined the
machine changes to high speed.
degree of improvement that can be achieved are
effected once.
The performance parameters are considered for
carrying out the analysis of the system is discussed
below.

 Level of limiting Machine Efficiency:

determining the average number of bottles
processed in the filling machine, which is the
limiting resource online, and is made by
dividing this number and bottles should be
Figure 27: 2D SIMIO model (unpacker machine processed in time modeling at the speed that is
properties and accumulation monitor charecteristics) the filler,
 Effective Efficiency Indicator Global:
2.4.13. Model Verification and Validation determining the average number of bottles
In order to perform a verification of the simulation processed and is performed by dividing this
model developed, a detailed analysis of each packaging number and bottles that should be processed in
process operations was accomplished. This assures us time modeling at the speed that is the
that model logic properly represents the sequence of palletizing machine,
operations of the real process.  Load Factor of transport: checks during each
In addition, the model validation executes an work shift the percentage of occupation rate
iterative comparison with the real system, making the each conveyor involved in the process in order
necessary adjustments and changes in the model until a to modify those that have a high occupancy
satisfactory similarity is achieved. and gain stability in the entire packaging line,

Proceedings of the European Modeling and Simulation Symposium, 2013 557

978-88-97999-22-5; Bruzzone, Jimenez, Longo, Merkuryev Eds.
  Number of pallets entered and graduated from 44.000
the line: the purpose of this parameter is to 42.000
determine the level of productivity that reaches 40.000

Speed (bpm)
the line under study for a work shift of eight
38.000
hours,
36.000
 Changes in speed and stability of the
machines: they want to reduce machine 34.000
downtime seeking stability thereof thus 32.000
explores the reasons why change their speed 30.000
(or shortage of product accumulation) and
proposes improvements.
Machine
Theoretical Speeds Real Speeds
3. RESULTS
The original design of line speeds under study is based Figure 28: "V" line with ideal and actual speeds
on the concept of the "V" which takes the limiting speed
of the machine, i.e. the filler, as a reference for defining Furthermore, the company determines the
machine speeds and hind to the same. The procedure for productivity of the packaging line from the
this calculation is carried out to increase between 10% measurement of the efficiency of their equipment. This
and 15% speeds as they move away from the filler. is calculated from the values corresponding to the
Ideal speeds (assuming ideal speed machine bottleneck, number of bottles produced during an operating period
550 bpm) and actual speeds of the machines determined in relation to the theoretical amount of
(considering uptime, downtime and uptime internal) bottles that must have occurred during that period
over a month of work were considered. Table 1 and (Equation 1). The bottles theoretical amount calculated
Table 2 show the results obtained from the analysis of from the limiting speed of the line which, as mentioned
speeds detailed above. above belongs to the filling machine.

Table 1: Analysis of theoretical speeds

(1)

Theoretical Speeds
Percentage of
For this calculation, reports were consulted
Machines capacity
Machines production of 3 consecutive months, of which we
Speeds (bph) limitation
obtained the total production time and the volume
machine
produced in the same. Thus, Table 3 shows in greater
Depalletiser 43.260 40
detail the productivity of each month analyzed. From
Unpacker 40.170 30
efficiency values shown in the above table is obtained
Washer 35.535 15 in the same behavior and the present trend in time
Filling 30.900 0 (Figure 4). Therefore, we can determine that the line has
Labeler 35.535 15 a 66.76% average productivity.
Packer 40.170 30
Palletizer 43.260 40 Table 3: Productivity Data
Actually Theoretical
Table 2: Analysis of actual speeds Average
Month Week produced produced
Real Speeds efficiency
bottles Bottles
Percentage Percentage 1° 3.187.638 4.752.000
Machines 2° 3.079.862 4.752.000
of machine of capacity 1 67,58
Machines Speeds 3° 3.269.809 4.752.000
capacity limitation
(bph) 4° 2.985.746 4.752.000
preceding machine
1° 3.082.036 4.752.000
Depalletiser 40.440 -9,7 30,9 2° 3.041.290 4.752.000
Unpacker 41.820 8,9 35,3 2 66,55
3° 3.424.400 4.752.000
Washer 38.400 24,3 24,3 4° 3.054.792 4.752.000
Filling 30.900 0,0 0,0 1° 3.155.279 4.752.000
Labeler 36.000 16,5 16,5 2° 3.069.707 4.752.000
Packer 38.160 6,0 23,5 3 66,17
3° 3.245.196 4.752.000
Palletizer 40.740 6,8 31,8 4° 3.108.242 4.752.000
The data in the table are expressed in Figure 28 for In this way, the efficiency of the model created can
a better visualization. It can be seen that the concept of be compared with the one of real system. Actually the
the "V" approaches the ideal nearby machines in the company has a line efficiency of 66.77%, which is
filler, but not at the ends.

Proceedings of the European Modeling and Simulation Symposium, 2013 558

978-88-97999-22-5; Bruzzone, Jimenez, Longo, Merkuryev Eds.
 similar to the obtained by the simulator because it Consequently, a 11.4% increase can be achieved in
reports a 66.8% value. efficiency by implementing minor changes without
In addition, the following productivity indicators incurring large investments while achieving significant
were used in order to validate the simulation model improvements in the operation of the line associated
developed: (i) number of pallets produced by shift (in with an increase in company profits.
both the feed and outlet line) and (ii) production in each It is noteworthy that the developed simulation
machine. model can be easily used for the evaluation of
The simulation model developed was executed alternative scenarios, i.e. the analysis of proposals for
several times to obtain performance mean values. The possible future changes in the design and operation of
information obtained is then compared with historical the line.
data of the line. This comparison is showed in Table 4.
4. DISCUSSION
Table 4: Performance measures of the real system vs the The decision variables allowing increasing both the
simulation model efficiency of the bottleneck machine and the efficiency
Real System Simulation of the overall line are the speeds of the machines that
Machine (pallets per Model (Pallets make up the transport and packaging process as well as
turn) per turn) the logic conditions programmed in the units.
Despalletiser 283 282 Furthermore, it is considered that these efficiencies
Empty bottle are less sensitive against increasing line capacity
280 278 drawer, which is associated with an accumulator device
inspector
Filling 277 275 for drawers or operator responsible for the same task.
Labeler 272 268 Short stops primarily derived from simple causes
Packer 278 274 can be reduced drastically without complex operations
Palletiser 271 270 on the machines, although there are also small stalls that
can only be removed using sophisticated methods of
Having analyzed all data from many runs, we analysis and operations with high technical content.
conclude that the simulation model developed has The causes that affect the productivity of the
acceptable apparent validity. packaging line, according to the simulation model
Having analyzed the most relevant scenarios, carried out, is the modification of the logic of the
presented above, key performance measures achieved in carriage of the feeding of the packaging process and
each of them are summarized in Table 5 and Table 6. clean room. Furthermore, the line is sensitive to changes
Therefore, it is possible concluding that scenario 5 in the speeds of the machines, which are operating at a
achieves the highest level of efficiency in terms of the speed below the nominal speed.
bottleneck resource and also the highest level of overall It is noteworthy that for fixed values of speed and
effective efficiency. This results in a remarkable transport machines, no investment is needed by the
increase in the production of a rolling line and the use company, because they have the materials and labor
of machines and transports. necessary for the modification of the same drivers.
Moreover, the study remarks that not always
Table 5: Summary of results obtained increasing the efficiency ratio on a particular machine
Processed Processed in Processed in line, from the reduction of a kind of loss, produces an
Scenario in filling bottles bottles increased rate of overall line efficiency. This is because
bottles depalletiser Palletizer the relationships and interactions in the real system are
1 157.200 151.800 complex or some degree of uncertainty is present.
165.059
It has been essential to have the automatic
2 161.375 160.200 154.200 registration of faults, without which no one could have
3 161.512 162.000 155.400 calculated the time lost in stops. On the other hand, it
4 177.462 178.800 175.200 would be beneficial to all line stoppages would be
5 206.200 211.200 204.000 assigned automatically and easily exported to a
spreadsheet. This would avoid much preparation work
Table 6: Summary of efficiency indicators and subsequent data analysis.
Percent Effective Global
Scenario ACKNOWLEDMENTS
Efficiency Efficiency
Financial support received from ANPCyT under grant
1 66,8 61,4
PICT-2010-1073, from CONICET under Grant PIP-
2 61,1 58,4 2221 and from UNL under Grant PI-66-337 is fully
3 61,2 58,9 appreciated.
4 67,2 66,4
5 78,1 77,3

Proceedings of the European Modeling and Simulation Symposium, 2013 559

978-88-97999-22-5; Bruzzone, Jimenez, Longo, Merkuryev Eds.
 REFERENCES Walpole Ronald, Myers Raymond, 1989, “Probability
Banks J., Carson J.S., Nelson B.L., Nicol D.M., 2004, and Statistics for Engineers and Scientists”, 4th.
"Discrete-Event System Simulation”; 4th. Ed.; Ed., Macmillan, N.Y.
Prentice-Hall U.S.A.
Christel, M.; Kang, K., 1992, “Issues in Requirements AUTHORS BIOGRAPHY
Elicitation”, CMU/SEI-92-TR-012, Pittsburgh,
EE.UU., Software Engineering Institute, Carnegie NATALIA BASAN is an Industrial Engineer and PhD
Mellon University. student conducting research in hybrid optimization &
Fishman George S., 1978, “Conceptos y Métodos en la simulation tools for production planning and scheduling
Simulación de Sistemas Discretos”. Limusa of automated production systems.
México.
Gordon Geoffrey, 1981, “Simulación de sistemas”, 2da. LUCILA RAMOS is an Industrial Engineer
Ed., Prentice Hall, México. conducting research in hybrid optimization &
Hepp, P.K.; “Introducción a la Adquisición de simulation tools for production planning and scheduling
Conocimiento para Sistemas Expertos”. of automated production systems.
Hines W. W., Montogomery D. C., 1990, “Probability
and Statistics in Engineering and Management MARIANA COCCOLA is an Information Systems
Science”, 3rd. Ed., Wiley, New York. Engineer and a PhD student conducting research in
Kelton D., Sadowski, R.P., Sturrok, D.T. 2006, hybrid optimization & simulation tools for production
"Simulation With Arena" Fourth edition. McGraw- and logistics systems.
Hill series in Industrial Engineering and
Management Science. Dr. CARLOS A. MENDEZ is a Titular Professor of
Kuo Benjamin; 1975, “Automatic Control Systems”; Industrial Engineering at Universidad Nacional del
3rd Ed.; Prentice-Hall, U.S.A. Litoral (UNL) in Argentina as well as a Researcher of
Law Averil, Kelton W. David; 1991, “Simulation the National Scientific and Technical Research Council
Modeling and Analysis”; 2nd. Ed.; McGrawHill, (CONICET) in the area of Process Systems
Inc., U.S.A. Engineering. He has published over 150 refereed
Mazzieri G., Obeid Z., 2004 “Distribución de journal articles, book chapters, and conference papers.
Mercaderías en Áreas Urbanas desde un Depósito His research and teaching interests include modeling,
Central”. Dpto. de Ing. Industrial, Fac. de Ing. simulation and optimization tools for production
Química, Univ. Nac. del Litoral. planning and scheduling, vehicle routing and logistics.
Musselman K, 1998, “Guideliness for Success” en
“Handbook of Simulation. Principles,
Methodology, Advances, Applications and
Practice”. Banks J. Editor. John Wiley & Sons,
Inc., USA.
Naylor Thomas, Balintfy Joseph, Burdick Donald, Chu
Kong; 1991, “Técnicas de simulación en
computadoras”; Editorial Limusa, México.
Papoulis Athanasios, 1990, “Probability and
Statistics”, Prentice Hall, Englewood Cliffs, N.J.
Piattini M., Calvo-Nanzano J., Cervera J., Fernández L.,
1996, “Análisis y Diseño Detallado de
Aplicaciones Informáticas de Gestión”. R. A. M.
A. Editorial, Raghavan, S.; Zelesnik, G.; Ford, G.:
“Lecture Notes on Requirements Elicitation”,
CMU.
Renee Thiesing, Christine Watson, Judy Kirby, David
Sturrock, 2009, “SIMIO Reference Guide”.
Versión 2.0.
Ross Sheldon M., 1993, “Introduction to Probability
Models”, 5th. Ed., Academic Press, N. Y.
Sapag Chain N., Sapag Chain R. 1989, “Preparación y
Evaluación de Proyectos de Inversión”. Mc.
Graw-Hill.
Senn, J.A., 1988, “Análisis y Diseño de Sistemas de
Información”, McGraw-Hill.
Shannon, Robert; 1988, “Simulación de Sistemas.
Diseño, desarrollo e implantación”; Trillas,
México.

Proceedings of the European Modeling and Simulation Symposium, 2013 560

978-88-97999-22-5; Bruzzone, Jimenez, Longo, Merkuryev Eds.