Proceedings of the World Congress on Engineering 2011 Vol I

WCE 2011, July 6 - 8, 2011, London, U.K.

Optimization of a Multiphase Multiproduct

Production Line based on Virtual Cells
Luís Pinto Ferreira, Enrique Ares Gómez, Gustavo Peláez Lourido, Benny Tjahjono
having to interrupt the normal functioning of the system
Abstract—The work presented in this article consists of the [1, 6].
development of a discrete event simulation model, based on a • Simulation allows one to identify the blocking and
real case and implemented in an Arena simulation environment. starvation points on the production line, as well as test
This is aimed at a class of very specific multiphase multiproduct the various options with a view to optimizing their
production lines, with a four closed-loop network
performance, thus identifying the causes for delays in the
configuration, which is very common in the automobile sector.
Also included in this study is an index of economic and flow of materials, information and other processes
functional efficiency, which evaluates cost-efficiency in each of [1, 5, 7].
the simulated scenarios. From a conceptual perspective, the • It can be used for testing explanatory hypotheses as to
idea of a virtual cell, was used in the configuration modeling of how and why a specific phenomenon occurs in the
the production line. The modeling of complex production system [2].
systems, based on the concept of virtual cells, allows for a • Simulation allows one to study a system which
greater understanding of the functioning process of a encompasses a wide time-span in a compressed period of
production line, through the simplification of the production
flow and the addition of buffers between the virtual cells. Thus,
time, or alternatively, it allows one to make a detailed
due to the facility presented by their dynamic configuration, study of the system’s performance over an extended time
virtual cells constitute an important modeling tool, which period [8].
should always be used whenever one aims to optimize • It can be used to know the system better and to identify
production flow, namely in the case of production lines with a which variables really influence its performance, acting
rather complex configuration. as a support for an improved understanding of reality so
as to explain and describe it, or as a support in
Index Terms— Virtual Cells, Simulation, Automobile decision-making [9, 10].
Assembly Lines
• It also enables users to test the system’s behaviour when
confronted with new and unexpected situations [1, 10],
I. INTRODUCTION AND LITERATURE REVIEW as well as constituting a support for decision-making
Simulation is one of the tools of support in the regarding investment in new technologies and
decision-making process, which allows for the projection equipment, improving production capacity and the
and analysis of performance in systems and complex management of material and human resources [11].
processes. Simulation can be seen as the construction Despite all the benefits one can gain from the use of
process of a model representing a real system, as well as that simulation as a support tool in the decision-making process,
of undertaking experiments using the same model with the it also presents a few relevant drawbacks, the most
purpose of better understanding its behaviour and evaluating significant of which are:
the impact of alternative operational strategies [1, 2, 3, 4]. • Simulation does not provide optimal solutions for case-
Many of the advantages ensuing from the use of study problems - it does, however, allow one to evaluate
simulation can be found in pertinent literature in this field. the system’s behaviour in accordance with specific
Some of these advantages can be clearly stated as follows: scenarios, which are created by the analyst for the
• It permits production engineers to test new purposes intended [1,2].
configurations in the production process, without the • If the model does not constitute a valid representation of
need to compromise resources, whose costs would be the system being studied, the simulation results will not
high [5]. produce much useful information about the real system
• It can be used to explore new resource grading policies, [8, 10].
operational procedures, decision-making rules, The work described in this article consists of the
organization structures and information flow, without development of a simulation model in an Arena
environment, based on a real case, and is aimed at a very
specific class of production lines, with a four closed-loop
Manuscript received February 3, 2011; revised February 28, 2011.
Luís Pinto Ferreira is with the Escola Superior de Estudos Industriais e de network configuration, commonly used in the automotive
Gestão (Technical-Scientific Unity of Industrial Engineering and sector. This study is a sequence of another, developed by
Production), Instituto Politécnico do Porto, Portugal (e-mail: Resano et al. [12,13,14,15], who designed one of the first
Luispintoferreira@eu.ipp.pt).
analytical models for an assembly line in the vehicle sector
Enrique Ares Gómez is a Full professor in Vigo University, Área Ingeniería
de los Procesos de Fabricación, Spain (e-mail: enrares@uvigo.es). as a network of four closed loops of machines, decoupled by
Gustavo Peláez Lourido is with Vigo University, Área Ingeniería de los intermediate buffers formed by conveyors. They consider
Procesos de Fabricación, Spain (e-mail: gupelaez@uvigo.es). that machines process pallets which are not univocally
Benny Tjahjono is with Cranfield University, School of Applied Sciences, related to each other. Both in the analytical model, as well as
Manufacturing Department, UK (e-mail: B.Tjahjono@cranfield.ac.uk).
in the simulation presented here, one can analyse the

ISBN: 978-988-18210-6-5 WCE 2011

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)
Proceedings of the World Congress on Engineering 2011 Vol I
WCE 2011, July 6 - 8, 2011, London, U.K.

blocking and starvation phenomena on a complex lots of parts. According to [27], the greatest difference
production line; these models also consider the proportion of between a virtual cell and the traditionally manufactured cell
four- and two-door cars between the door disassembly and resides in the dynamic nature of the former. In the case of
assembly stations. There is no knowledge of any other work traditionally manufactured cells, the location of work
in specialized literature, in the area of simulation, which stations is fixed and perfectly identifiable. The same does
includes all these factors, considered to be of such specific not happen with virtual cells, since they will vary according
relevance in the performance of these types of to the requirements at hand. According to [28], a virtual cell
manufacturing systems [16,17,18]. is any production scenario, from the lowest levels in a
There are, in fact, very few studies undertaken which use manufacturing hierarchy (for example, a CNC machine), to
simulation as a support tool in the decision-making process the highest (for example, a factory). Its internal organization
in the context of networks of closed-loop production lines. allows for the manufacture of different types of products, as
There is, however, a wider range of bibliography whenever well as that of establishing significant parameters in its
the purpose of the study presupposes the development of production capacity.
analytical models for these types of configurations. [19] Figure 1 represents the main automobile line, as well as
developed a simulation model based on the Siman language, the door and front axle preassembly lines, as a network of
with the purpose of maximizing the production line rate, four closed loops of machines (M) and intermediate buffers
with a closed-loop configuration, through the optimization (B) formed by conveyors. Each of the intermediate buffers
of time distribution over the entire cycle, in all the different possesses a limited capacity, in accordance with the number
work stations. [20] developed a model for a network of pallets it can contain. The time cycle for each machine is
configuration to be used on a production line, comprising fixed and deterministic.
various machines, where assembly and disassembly
operations are undertaken. This model is based on the first M5 B5,6 M6
simulator for continuous flow, introduced by [21], where the
discrete traffic of components is likened to a continuous
M7
flow. However, this study does not consider the phenomena B2,5 B6,4
of blockage transmission and starvation in a network of B3,7 B7,3
various closed loops. [22] proposed an analytical model to
estimate the performance parameters of a closed-loop
production line, characterized by limited capacity buffers M1 B1,2 M2 B2,3 M3 B3,4 M4
and failure-prone machines. [23] resorted to the “Quest”
language in order to develop a simulation model for a B4,2
production line with a closed-loop configuration. Their
purpose was to optimize its functioning through a better B4,1

balance in the production flow. In fact, various rules and Figure 1. Main automobile assembly line and preassembly lines.
conditions were evaluated so as to ensure a more effective
planning and control of the entire production system. Yet, The sum of the car bodies, door and front axle assembly
for this study, the authors did not take into account the pallets stored in each of the intermediate buffers of the three
phenomena of machine blockage and starvation. Neither did first closed-loops remain, respectively, constant at any time
they consider an external variable (x). These factors were and are defined by Equations (1) to (3). This fact, as well as
included in the simulation model developed in this study. other analyses undertaken to evaluate the influence of
[24] created a tool to be used in the analysis of production different variables which interfere in the performance of the
lines with a network configuration of various closed loops. automobile assembly line being studied, was previously
He developed a graphic and induction model to analyze analysed in [12, 13, 14, 15].
efficiently the phenomena of blockage and starvation.
Nevertheless, for the construction of the model, this author n 12 + n 23 + n 34 + n 41 = 237 (1)
did not introduce an external variable (x), which is n 25 + n 56 + n 64 + n 42 = 450 (2)
represented by the four-door car body proportion in the
simulation model developed within the framework of this n 37 + n 73 = 138 (3)
study. [25] developed an analytical model to evaluate and
optimize the performance of an assembly line of The fourth closed-loop defines the relationship between
semiconductors, characterized by a two closed-loop the number of pallets with doors in different preassembly
configuration. This study aimed to evaluate the impact of states and the number of pallets of cars with disassembled
several parameters on the performance of the system, such as doors via an external variable (x), according to Equation (4).
the number of circulating pallets and machine processing This variable represents the four-door car ratio between the
times, amongst others. doors’ disassembly stations, located at M 2 , and the
doors’assembly stations, located at M 4 . This variable can
II. CONCEPTUAL MODELING BASED ON VIRTUAL CELLS take on values of between 0 and 1.
From a conceptual perspective, one used the concept of a
virtual cell in the modeling of the production line. This cell Equation (4):
is also known as a reconfigurable cell. The concept of the n 25 +n 56 +n 64 +n 5 +n 6 = (n 2 +n 3 +n 4 +n 23 +n 34 )·(1 + x )
virtual cell was first introduced by [26] in order to develop
control software for an automated production line of small

ISBN: 978-988-18210-6-5 WCE 2011

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)
Proceedings of the World Congress on Engineering 2011 Vol I
WCE 2011, July 6 - 8, 2011, London, U.K.

Where n 5 and n 6 are the numbers of stations M 5 and Virtual cell 2

M 6 , n 2 is the number of stations at M 2 , from the door M5 B5,6 M6

disassembly stations, n 3 is the number of stations at M 3 ,
and n 4 is the number of stations at M 4 to the door M7
B2,5 B6,4
assembly stations. The values n 2 , n 3 , n 4 , n 5 and n 6 remain
B3,7 B7,3
constant. Replacing the sum of their values in Eq. (4) the Virtual cell 3

following is obtained [12, 13, 14, 15]:

M1 B1,2 M2 B2,3 M3 B3,4 M4
n 25 +n 56 +n 64 +60 = (216 +n 23 +n 34 )·(1 + x ) (5)
Virtual cell 1
B4,2

The simulation model presented in this article is based on B4,1

these equations in order to determine the number of pallets
which must circulate on each of the 4 closed-loops which Figure 3. Modeling of the automobile assembly line based on 3 virtual
exist on the analyzed automobile assembly line. cells.
On the first three loops, the velocity (V) of the intermediate
buffers formed by conveyors is: V1,2=16m/minute, Another modeling example based on the concept of
V2,3 = 16m/minute, V3,4 = 16m/minute, V4,1 = 16 m/minute, virtual cells is presented in figure 4, in which the automobile
V2,5 = 25m/minute, V5,6 = 25m/minute, V6,4 = 25 m/minute, assembly line is structured on the basis of 2 virtual cells.
V4,2 = 25m/minute, V3,7 = 45m/minute, V7,3 = 45m/minute.

Figure 2 presents the automobile assembly line analyzed M5 B5,6 M6

for this study. In the same manner as [12], 7 virtual cells
were constituted so as to better understand the trajectories of
M7
the state of the intermediate buffers formed by conveyors, B2,5 B6,4
positioned between each virtual cell and thus ensuring B3,7 B7,3
supply to the cells and avoiding starvation points. These
virtual cells are represented by each of the machines (M)
presented in figure 2. In the automobile production line M1 B1,2 M2 B2,3 M3 B3,4 M4
being analyzed, each of these constitutes a set of different
types of stations (for example, manual stations, automatic B4,2
stations, semi-automatic stations, etc.), which are logically
grouped with the purpose of undertaking a specific set of B4,1

tasks. For this study, one used this modeling of line Figure 4. Modeling of the automobile assembly line based on 2 virtual
configuration to develop a decision-making support system cells.
based on a discrete event simulation model for a multiphase
multiproduct automobile production line. The main III. CHARACTERISTICS OF THE DECISION-MAKING SUPPORT
objective here was to compare the results obtained by [12], SYSTEM DEVELOPED
which is based on analytical methods, with those obtained In the context of this project, all the work was developed
through the simulation models. in an Arena simulation environment [29]. Its simulation
Virtual cell 5
language constitutes a visual and flexible programming tool,
Virtual cell 6
M5 B5,6 M6
directed at the object, since it simultaneously combines the
construction of simulation models with the integration of
Virtual cell 7
different commonly used languages: Visual Basic, C, C++.
M7 This language is based on the SIMAN simulation language
B2,5 B6,4
B3,7 B7,3
[30].
Virtual cell 1 Virtual cell 2 Virtual cell 4
The purpose of the use of an Arena simulation
environment was that of enabling the production engineer to
M1 B1,2 M2 B2,3 M3 B3,4 M4
evaluate the performance of the automobile assembly line,
Virtual cell 3 through the variation of different parameters, thus
B4,2
contributing to an improved specification, characterization
B4,1
and definition of the most efficient control system. With this
objective in mind, a support system for decision-making was
Figure 2. The modeling of the automobile assembly line based on 7 virtual developed; this enables the automatic generation of different
cells.
simulation models. In the initial stage of simulation, the user
is able to interact with the system to be developed, through
Another line configuration modeling option based on 3
the introduction of various parameters such as:
virtual cells could be used for the analysis and optimization
of the automobile assembly line being studied, as is • The four-door car ratio (x);
presented in figure 3. • The processing time for each machine;

ISBN: 978-988-18210-6-5 WCE 2011

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)
Proceedings of the World Congress on Engineering 2011 Vol I
WCE 2011, July 6 - 8, 2011, London, U.K.

• The production sequence in accordance with each type

of car (two- or four-door);
• The speed and length of the intermediate buffers formed
by conveyors;
• Simulation time;
• The number of pallets circulating on the first three
closed-loops.

Figure 5 presents the graphic interface which allows the

user to configure the above mentioned parameters.
Figure 7. Number of automobiles of each type produced.

Figure 8. Automobile production time.

Figure 5 – Parameterization of the automobile assembly line.

Another contribution made by this study was that of IV. VALIDATION OF THE SIMULATION MODEL PROPOSED
integrating an automatic report generator in the simulation In [12, 14, 15] the incompatibility between equations (2)
environment; this presents information on the main and (5) (see section 2) was demonstrated. These define the
performance indexes of the simulated models, using graphic sum of the number of pallets on the intermediate buffers
visualization, which required the integration of various which constitute the second and fourth closed loop, with
applications in the Arena simulation environment, such as: some of the maximum and minimum capacities of the
Visual Basic for Applications (VBA), Visual Basic (VB), corresponding intermediate buffers, for the specific values of
Microsoft Access (Database), Crystal Report (Graphics variable x. This incompatibility reveals that the automobile
Edition). In this context, it is worth highlighting the visual assembly line cannot operate in practice for x<0.37, in both
aspects of the graphics produced (three examples are the stationary and transitory regimes, and for x>0,97, in the
presented in figures 6, 7 and 8), which allow the user to have stationary regime. As with any simulation project, an
a better perception of the performance of the simulated important phase is the validation of the model constructed;
models. this ensures that it is a faithful representation of reality and
From among those considered, the following performance can then be used for decision-making. Through the
indicators are highlighted: simulation model developed in the context of this study, and
• Machine Usage Levels; for the automobile production line being analyzed, one also
• Number of Cars for each of the types produced; verified that the system did not function for the values of
• Number of operations undertaken on each of the x<0,37 and x>0,97, due to the phenomena of blockage and
machines; starvation occurring on the machines integrating the
• Production time/vehicle; analyzed model. In the same way, one also noticed that, once
• Relationship between the time cycle of each machine the minimum buffer capacity mentioned in [12, 13] was
and the number of pallets on each upstream buffer. reached, a continuous supply of transport and assembly
pallets to machines was assured [16]. In the context of this
work, the validation of the simulation model proposed was
confirmed, on the basis of these two presuppositions.

V. INDEX OF ECONOMIC AND FUNCTIONAL EFFICIENCY

It is important to quantify the benefit obtained whenever
various parameters are altered in the multiphase
multiproduct production line being studied. One is then able
to quantitatively assess the efficiency of each of the
simulated scenarios. One sought to develop an indicator of
an economic type which, for a given period of time, would
calculate the most representative costs for each of the
Figure 6. Number of operations undertaken on each of the machines.
simulated alternatives. Thus, based on the economic function
index developed by [31, 32, 33, 34], a new economic
function index (EFI), was elaborated, which considered the

ISBN: 978-988-18210-6-5 WCE 2011

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)
Proceedings of the World Congress on Engineering 2011 Vol I
WCE 2011, July 6 - 8, 2011, London, U.K.

specific aspects pertaining to the multiphase multiproduct VI. CONCLUSION

automobile production line being studied. Table 1 below In this paper, an automobile assembly line and different
presents the variables used in the development of the (EFI) preassembly lines were modelled as a network of four
so as to evaluate its cost-effectiveness. closed-loop machines and intermediate buffers formed by
conveyors, constituting a configuration which is widely used
TABLE 1 in these kinds of assembly lines. Using an Arena simulation
VARIABLES USED IN THE DEVELOPMENT OF THE EFI
environment, a representative model of this line was
Nij Total number of type i parts produced for each j machine
PTij Processing time for the type i part on machine j
developed with the purpose of providing the production
MCij Manufacturing cost of the type i part on machine j engineer with a better understanding and assessment of its
per time unit performance; it represents, on the computer screen, each
CNUj Cost of non-usage of machine j per time unit moment in the state of the system and its evolution over
TIj Time of inactivity of machine j time, thus allowing for improved communication between
CPIFD Costs of penalties incurred due to failure to meet delivery
deadlines
the model and the user. In fact, the great visual component
TDO Time of delay in completing an order of the models developed allows for a clear understanding on
CPz Cost based on the number of pallets circulating on loop z the part of the decision-making agents who will eventually
(z =1,2,3) use them, since they will be able to view the interaction
ECLz Energy costs for loop z (based on the loop’s velocity, the occurring between the entities which make up the model,
number of pallets, etc.) (z =1,2,3)
CLCLz Cost associated to the length of the conveyors on loop z
allowing for a better understanding of the simulation results.
(z =1,2,3) The authors believe that, by producing this study, they have
contributed to emphasising the importance of the use of
All the variables are real and the costs are obtained on the simulation in the development, on computer, of virtual
basis of a pre-established time unit. The EFI includes the environments which reproduce the behaviour of many real
following components: systems.
As perspectives of future work, some actuation lines stand
EFI = Fixed cost of manufacture + Cost of machine out, as a natural continuation of the proposals presented
non-usage + Cost of failure to meet delivery deadline + Cost here:
of pallets + Energy costs + Cost associated to the length of • Evaluated the impact of the pallet numbers circulating on
the conveyors the first three closed loops on the performance of the
production line, translated into the number of cars
In this manner, and considering the previously described produced/hour, in order to improve the availability of the
variables, the following economic index function was entire manufacturing system for any value of x.
obtained. This will allow one to calculate the most • Assess the impact of the length of intermediate buffers on
representative costs for each of the simulated scenarios: the performance of the production line, translated into the
7 7 number of cars produced / hour, in order to improve the
EFI= ∑ Nij * PTij * MCij + ∑ CNUj * TIj + availability of the entire manufacturing system for any value
of x.
j =1 j =1
3 3 3 • Impact evaluation that the sequence of production of
CPIFD * TDO + ∑ CPz + ∑ ECLz + ∑ CLCLz different types of cars in the automobile assembly line
performance.
z =1 z =1 z =1
In the same way, one can calculate the most representative • Behaviour evaluation of the production line when a new
costs for each of the simulated scenarios, in accordance with car type is inserted.
the first three loops on the production line. Thus,
considering that EFI 1, EFI 2 and EFI 3 respectively REFERENCES
represent the economic índex function associated to loop 1, [1] Robert E. Shannon, “Introduction to the Art and Science of
loop 2 and loop 3, one can then obtain the following Simulation”, in Proceedings of 1998 Winter Simulation Conference.
[2] Ricki G. Ingalls, “Introduction to Simulation”, in Proceedings of
economic indicators: 2001 Winter Simulation Conference.
4 4 [3] Michael Andersson, Göte Olsson, “A Simulation Based Decision
EFI1= ∑ Nij * PTij * MCij + ∑ CNUj * TIj + Support Approach For Operational Capacity Planning In A Customer
Order Driven Assembly Line”, in Proceedings of 1998 Winter
j =1 j =1
Simulation Conference.
[4] Villa, A; Ares Gómez, J. E. ,1998, “A methodology to analyze
CPIFD * TDO + CP1 + ECL1 + CLCL1 workshop lines by discret event dynamic models”. Recent
Developments in Production Research, edited by A. Mital.
[5] Jerry Banks, “Introduction to Simulation”, in Proceedings of 2000
EFI2= ∑ Nij * PTij * MCij + ∑ CNUj * TIj + Winter Simulation Conference, J.A. Joines, R. R. Barton, K. Kang,
j = 2, 4,5,6 j = 2, 4,5,6 and P. A. Fishwick, eds.
[6] Ares, E., Peláez, G., “Integración de línea y célula flexibles en un
CPIFD * TDO + CP2 + ECL2 + CLCL2 modelo jerárquico para la planificación y programación de la
producción”, Febrero 1997, XII Congreso Nacional de Ingeniería
Mecánica, Anales de Ingeniería Mecánica, Revista de la Asociación
EFI3= ∑ Nij * PTij * MCij + ∑ CNUj * TIj +
[7]
Española de Ingeniería Mecánica.
Agostino Villa, Giovanni Fiorio, J. Enrique Ares Gomez, 1987,
j = 3,7 j = 3,7 “Production planning and control in multi-stage multi-product
systems”, Modern Production Management Systems, A. Kusiak
CPIFD * TDO + CP3 + ECL3 + CLCL3 (Editor), Elsevier Science Publishers B. V. (North-Holland),
ISBN: 0444702725.

ISBN: 978-988-18210-6-5 WCE 2011

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)
Proceedings of the World Congress on Engineering 2011 Vol I
WCE 2011, July 6 - 8, 2011, London, U.K.

[8] Guimarães Rodrigues, Valério de Carvalho, 1984, “CAPS – ECSL, [20] Kouikoglou VS., 2002, “An Efficient discrete event model of
Experiência de modelagem e simulação aplicada a um sistema de assembly/disassembly production networks”, International Journal of
elevadores”, Universidade do Minho. Production Research, 2002, 40(17):4485-4503.
[9] Averill M. Law, W. David Kelton, 1991, “Simulation Modeling & [21] D`Angelo H, Caramanis M, Finger S, Mavretic A, Phillis YA,
Analysis”, Second Edition, McGraw-Hill International Editions, Ramsden E., 1988, “Event-driven model of unreliable production
ISBN 0-07-036698-5. lines with storage”, International Journal of Production Research,
[10] Reuven Y. Rubinstein, Benjamin Melamed, 1998, “Modern 26(7):1173-1182.
Simulation and Modeling”, Wiley Series in Probability and Statistics, [22] David Douard, Bruno Baynat, “Improving Efficiency of
Applied Probability and Statistics Section, A Wiley-Interscience Decomposition Methods for Closed-Loop Continuos-Flow Production
Publication, John Wiley & Sons, INC., ISBN 0-471-17077-1. Models”, ACS`02 – SCM Conference, Poland, October 23-25, 2002.
[11] Charles Mclean, Swee Leong, “The expanding Role of Simulation in [23] Masayuki Soma, Shinsuke Kondoh, Yasusi Umeda, “Simulation of
Future Manufacturing”, in Proceedings of 2001 Winter Simulation Closed-loop Manufacturing Systems focused on Material Balances”,
Conference. in Proceedings of EcoDesign2003: Third International Symposium on
[12] A. Resano Lázaro, 2007, “Análisis Funcional y Optimización de la Environmentally Conscious Design and Inverse Manufacturing,
Distribuición en Planta de una Línea de Ensamblaje de Automóviles”, Tokyo, Japan, December, 8-11, 2003.
PhD Thesis, directed by Prof. C. J Luis Pérez. Departamento de [24] Zhang Z., 2006, “Analysis and design of manufacturing systems with
Ingeniería Mecánica, Energética y de Materiales”, Universidad multiple-loop structures”, PhD Thesis, Massachusetts Institute of
Pública de Navarra. Technology.
[13] A. Resano Lázaro, C. J. Luis Pérez, 2008, “Analysis of an [25] Na Li, Li Zheng, Quan-Lin Li, “Performance Analysis of Two-Loop
automobile assembly line as a network of closed loops working in Closed Production Systems”, Computers & Operations Research,
both, stationary and transitory regimes”. International Journal of Volume 36, Pages: 119-134, 2009, ISSN: 0305-0548, Elsevier
Production Research, 1366-588X, Volume 46, Issue 17, Science Ltd.
Pages 4803-4825. [26] C. R. Mclean, H. M. Bloom, T. H. Hopp, “The Virtual
[14] A. Resano Lázaro, C. J. Luis Pérez, 2007, “Analysis and Manufacturing Cell”, 1982, Proceedings of Fourth IFAC/IFIP
enhancement of the four-door cars proportion limits in a real Conference on Information Control Problems in Manufacturing
automobile assembly line working in transitory regime”, Second Technology, Gaithersberg, MD, USA.
Manufacturing Engineering Society International Conference. [27] Adil Baykasoglu, “Capability-based distributed layout approach for
Madrid. Spain. virtual manufacturing cells”, 2003, International Journal of
[15] A. Resano Lázaro, C. J. Luis Pérez, 2007, “Dynamic analysis of an Production Research, Volume 41, Issue 11, pages 2597 – 2618.
automobile assembly line considering starving and blocking. [28] Ares Gómez E., Peláez Lourido G., Gómez Lourido R., “Modelo
Robotics and Computer Integrated Manufacturing”, Article in Press. Dinamico para la Evaluacion de la Eficiencia de Estrategias de
DOI:10.1016/j.rcim.2007.11.002. Produccion en Ambiente CIM”, Diciembre 1990, Anales de
[16] Luís Pinto Ferreira, E. Ares Gómez, G.C Peláez Lourido, A. Resano Ingeniería Mecánica, Revista de la Asociación Española de Ingeniería
Lázaro, C. J. Luis Pérez, “Comparative Analysis between analytical Mecánica.
and Arena simulation models applied to an Automobile closed-loops [29] W. David Kelton, Randall P. Sadowski, David T. Sturrock, 2007,
assembly line, modelled like a network of closed-loops”, in “Simulation With Arena”, Fourth Edition, McGraw-hill Series,
Proceedings of 25th International Manufacturing Conference: ISBN-13: 978-0-07-110685-6.
“Manufacturing and Design – The Next Generation” (IMC2008), [30] David A. Takus, David M. Profozich, “ARENA Software Tutorial”,
Ireland, 3rd-5th September 2008. in Proceedings of 1997 Winter Simulation Conference.
[17] Ferreira, L[uís]; E. Ares, G[ómez]; G.C Peláez, L[ourido] & Salgado, [31] Ares Gómez, J. E. “Estructura Jerárquica de Metodologías para la
M[arina] (2010). Analysis of the Influence of Conveyor Speed on the Implantación y Gestión de Sistemas de Fabricación Flexible”, Tesis
Behaviour of an Automobile Assembly Line, Chapter 41 in DAAAM Doctoral, 1986, Universidad de Santiago de Compostela.
International Scientific Book 2010, pp. 463-470, B. Katalinic (Ed.), [32] Ares Gómez J. E., Perez Garcia J. A. “Estrutura de sistema experto
Published by DAAAM International, ISBN 978-3-901509-74-2, ISSN en tambem mode para la toma de decisiones en el diseño e
1726-9687, Vienna, Austria. DOI:10.2507/daaam.scibook.2010.41. implantación de lay-out en un ambiente CIM”, Diciembre 1990,
[18] Luís Pinto Ferreira, Enrique Ares Gómez, Gustavo Peláez Lourido, Anales de Ingeniería Mecánica, Revista de la Asociación Española de
Marina Salgado, José Diéguez Quintas, “Analysis on the Influence of Ingeniería Mecánica.
the Number of Pallets Circulating on an Automobile Closed-Loops [33] M. A. Sebastián Pérez, J. Conde Collado, E. Ares Gómez, Agostino
Assembly Line”, January 2011, International Journal of Advanced Villa, “Enfoques acerca del diseño, implantación y optimización de
Engineering Sciences and Technologies (IJAEST), Volume number 2, sistemas de fabricación flexible”, Octubre de 1990, Publicaciones del
Issue number 2, pp. 119-123. VIII Congreso de Investigación, Diseño y Utilización de
[19] Rajan Suri, Ying Tat Leung, “Single Run Optimization of a Siman Máquinas-Herramienta, San Sebastián.
Model for Closed Loop Flexible Assembly Systems”, in Proceedings [34] Pérez Garcia J. A., Pelaez Lourido G., Ares Gómez J.E.,
of the 1987 Winter Simulation Conference, A. Thesen, H. Grant, W. “Transformación de un Taller Tradicional de Fabricación a una
David Kelton, eds. Filosofia CIM”, 1992, IX Congreso de Investigación, Diseño y
Utilización de Máquinas-Herramienta. Publicación de Inverna.

ISBN: 978-988-18210-6-5 WCE 2011

ISSN: 2078-0958 (Print); ISSN: 2078-0966 (Online)