Second Asia International Conference on Modelling & Simulation

Simulation and Re-engineering of Truck Assembly Line

Arya Wirabhuana, Habibollah Haron, Muhammad Rofi Imtihan
Department of Modeling and Industrial Computing
Faculty of Computer Science & Information Systems,
Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor, Malaysia
[email protected], [email protected], [email protected]

Abstract the impact of capital investments in equipment and

Simulation is one of tools that has been used widely in physical facility, and to assist in proposing changes to
several manufacturing areas and organizations as well material handling and layout. They have also found it
as in automotive industries. Using a valid simulation useful to evaluate staffing and operating rules, and
model may give several advantages in creating better proposed rules and algorithms to be incorporated into
manufacturing design in order to improve the system warehouse management control software and produc-
performances. This paper presents result of tion control systems. Furthermore, simulation is very
implementing a computer based simulation model to useful in providing a "test drive" before making capital
design Manufacturing Process Re-engineering for investments, without disrupting the existing system
performances improvement. The basic problem is that with untried changes.
the current manufacturing system performances have to This paper presents how simulation can be applied
be improved to deal with the business environment. into Manufacturing Process Re-engineering Design in
Projects objective is to design four improvement Automotive industries sub system in order to improve
alternatives design using valid computer simulation their performances using certain manufacturing
model. Three approaches applied as the foundation in performance improvement approaches. Each design will
creating improvement alternative of the real system are represent the solution alternative that is proposed based
Line Balancing, Facilities re-layout and process on particular manufacturing improvement technique.
enhancement, and manufacturing process. Simulation This paper is divided into five sections. Section 2
models were developed using ARENA version 7.1 while gives literature review and definition few terms, Section
Statfit and Microsoft Excel were used for statistical 3 presents the current system while Section 4 presents the
analysis. The case study was taken from the Job Shop simulation model and the analysis on the model. Section
Manufacturing line of Body Welding and Metal Finish 5 concludes the paper.
operations of truck assembly line. The result derived
from simulation model of current system and four 2. Literature Review
proposed simulation models that are based on three
approaches are analyzed and presented to be considered This section presents few definitions that related to
as new strategy in improving the truck assembly line. the discussion. Four definitions given are what is
simulation and its application in manufacturing industry,
1. Introduction definition of line balancing, facility layout, and
automated production system.
Simulation is a contemporary Information A system is defined as a collection of entities,
Technology (IT) based technology that was raised to usually people and machines, which act and interact
minimize risk and speed-up problems solutions. In order toward the accomplishment of some logical end [2].
to develop a valid simulation model, companies can Simulation, according to Shannon [3], is the process of
conduct experimental design to get a new design of their designing a model of a real system and conducting
manufacturing system without disturbing the working experiments with this model for the purpose either of
system. Computer simulation model accommodates understanding the behavior of the system or of evaluating
implementation of various Manufacturing Process Re- various strategies. Carson [4] stated that simulation is
engineering Designs into computer based model and most useful in seven situations; for example when there
simulate it as well as justifying the performances. is no simple analytic model that is sufficiently accurate to
Manufacturing and material handling systems analyze the situation.
provide one of the most important applications of The use of computer simulation in design and
simulation. Simulation has been used successfully as operation of automotive industries is very helpful
an aid in the design of new production and especially in car and truck assembly plants. Most of the
manufacturing facilities, warehouses, and distribution automotive manufacturers world-wide, such as Toyota
centers as well as in automotive industries. It has also [5], General Motors [6], Ford [7], and Mercedes-Benz [8]
been used to evaluate suggested improvements to currently require new and modified manufacturing
existing systems [1]. Generally, engineers and analysts system designs to be verified by simulation analysis
have found that simulation is very valuable to evaluate before they are approved for final manufacturing design.
An integrated approach of how simulation being used in

978-0-7695-3136-6/08 $25.00 © 2008 IEEE 783

DOI 10.1109/AMS.2008.82

Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on January 11, 2009 at 20:11 from IEEE Xplore. Restrictions apply.
 designing and evaluating a new manufacturing facility the design of equipment to accommodate a specific class
layout was show by Shady et al. [9]. of product changes, and modifying the control program
In this paper, the impact of a new work cell design can change the processing or assembly operations.
in improving manufacturing productivity is presented. Flexible automation refers to the design of equipment to
The work focused on an assembly line consisted of many manufacture a variety of products or parts and very little
separate and distinct work elements. The sequence in time is spent on changing from one product to another.
which these elements can be performed is restricted, at These concepts is important in determine the type of
least to some extent, and the line must operate at a automation of a manufacturing process.
specified production rate, which reduces to a required
cycle time. Given these conditions, the line-balancing 3. The Truck Assembly Line
problem is concerned with assigning the work elements to
workstations so that all workers have an equal amount of This section presents the current practice of the
work. The objective in line balancing is to distribute the truck assembly line and problems faced by the
total workload on the assembly line as evenly as possible management. The current manufacturing processes of the
among the worker [10]. Two important concepts in line truck assembly line consist of 12 main processes. First
balancing are the separation of the total work content into half of the processes are made up of two sequences. The
minimum rational work elements and the precedence first sequence is Body Shop, Metal Finish, Paint Shop,
constraints that must be satisfied by these elements. . and Trimming Cabin process while the second sequence
Facilities design for manufacturing systems is also are Rivet, Axle, Trimming Chassis and Engine Drop
important for a company because of the productivity process. These sequences are conducted in parallel and
dependence on manufacturing performance. Since combine at cabin drop section. The following processes
manufacturing is a value-added function, the efficiency are Final Assembly, Quality Control, Recty, and lastly is
of the manufacturing activities will make a major Delivery. Figure 1 shows model of the truck assembly
contribution to the company productivity. Greater line. Figure 2 shows the production flow process that is
emphasis on improved quality, decreased inventories, the detail of processes in Figure 1.
and increased productivity encouraged the design of
manufacturing facilities that are integrated, flexible, and
responsive. Tompkins et al. [11] wrote that the
Body Metal Paint Trimming
effectiveness of the facility layout and material handling Shop Finish Shop Cabin
in these facilities will be influenced by a number of
factors, including changes in product mix and design,
processing and materials technology, handling, storage, SUB ASSY
Cabin Final
Assembly
Drop
and control technology, production volumes, schedules,
and routings, and management philosophies. Tompkins et
al. [11] stated that the manufacturing plant or facilities Trimming Chassis Quality
layout might be differentiate into four types which are Rivet Axle & Control
fixed position layout, process layout, cellular/group Engine Droop

technology layout, and product layout.

Automation is the technology that the process or Recty
procedure is accomplished without human assistance.
It is implemented using a program of instructions
combined with a control system that executes the
Delivery
instructions. Groover [10] divides manufacturing
automation into three types namely Fixed,
Programmable, and Flexible automation. Fixed
automation refers to the use of custom-engineered
equipment to automate a fixed sequence of processing or Figure 1 : General truck assembly processes
assembly operations. Programmable automation refers to
The company grand total production capacity of whole also shows that there might be concluded that every
type is 15000 units per year, whereby the truck workstation seemed have different number of component
production is limited only about 6900 units annually. types to be assembled. Since the different number
Since the significant rise of product demand in the past 3 component types between every workstation is quite
years, so the company is facing the problem in improving significant, it can be presumed that there is possibility
the truck production capacity. From the current that every workstation might have different workload of
manufacturing layout, conclusion might be taken that each others, in another ways it can be presumed that the
there are still significant possibilities to improve the body welding operations might have balance workloads
current layout. Another problem that could be found in that drive less manufacturing line efficiency.
the current system is the significant discrepancies of There are four types of primary data collected for
periodic production capacity. It indicates that the current this project. Figure 3 describes the classification diagram
manufacturing system was not well standardized. Studies of the primary data. The first type is the Factory layout

784

Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on January 11, 2009 at 20:11 from IEEE Xplore. Restrictions apply.
 and dimension. These data are used as primary
consideration in developing a new proposed layout for
the improvement system. The second data type is
operation time. The operation time covers the
workstation operation processes and transfer time.
Manufacturing standard time (STT) parameter is used to
represent workstation operation time while transfer time
is divided into inter–workstation transfer time and inter-
section transfer time that describe the time required to
transfer work in process product from Body Welding
section to Metal Finish section. After that, the data will
be examined to ensure the validity to be used for
simulation model and model analysis. Data validity that
is based on these two types is data sufficiency and data Figure 8: Comparison of output standard
quality. The other two types of primary data that
collected are number of daily finish product or called The current model produced 29 products per day
Output standard and number of additional Work-in- while the first model only accommodates 28 units daily.
process in buffer area. Both data are used in the model Although the first model might possible have higher line
validation phase and to compare the system performance. efficiency but the production capacity is lower than the
current model. The confidence intervals for output
standard for each model are also compared. For second
Factory Facility Layout and Dimension model, the output standard increased dramatically until
Workstation Operation Time
= Standard Transfer Time 43.5 unit per day or around 50% higher than the initial
Operation Inter – Workstation model. This proved that parallel operation in Metal
Time Transfer Transfer Time Finish section could significantly improve the production
Time
Primary Body Welding – Metal capacity. The third model might give 6 additional
Data Output Standard Finish Transfer Time
(Daily Finish Product)
product daily or about 14% higher than the second
Will be used for model. With this proposed system, 49.5 unit of daily
Daily Additional WIP model validation production capacity can be achieved satisfactorily. The
in Buffer area
fourth model can boost the daily output standard at 56
units per day, which is 13% higher, that previous model
and more than 93% higher from the initial model. This
Figure 3 : Primary Data Collection result is also justified that process breaking down and
process automation might increase the production
capacity radically.
4. Simulation Model and Analysis
4.2 Product Cycle Time
This section presents four proposed simulation
models of the current truck assembly line and analysis of The second performance indicator is product cycle
the models based on three performance indicator. The time, which are the interval time between finish products
models are line balancing, parallel operation, separated created at the end of the manufacturing processes.
material handling system (MHS), and full synchronized Workstation standard time and transfer time are the most
MHS as shown in Figure 4, Figure 5, Figure 6 and Figure important elements in finish product cycle time
7 respectively. The performance indicators are output formulation. Figure 9 illustrated the comparison of the
standard, finish product cycle time, and manufacturing finish product cycle time for simulation and current
line efficiency. model. As stated in Figure 9, initial model cycle time is
22.5 minutes per product. The simulation models
4.1 Output Standard seemed could reduce the cycle time from first model
until fourth model that is: 20, 17.9, 9, and 8 minutes per
The first performance indicator is output standard product respectively. It means that the product cycle
which is a common normal daily production capacity of time totally might reduce until 60% by fourth model.
manufacturing line. Figure 8 shows comparison between Comparison of the first and second parameter
current and simulation model. shows few conclusions. Third model gave the most
significant improvement from 17.9 minute reduced
almost 50% to 9 minute per product. Since the second
model apply parallel processing while third model apply
automated process, so it can be determined that parallel
processing will improve significantly the production
capacity, but process automation will mainly effect the
cycle time. As shown in Figure 9, since automated

785

Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on January 11, 2009 at 20:11 from IEEE Xplore. Restrictions apply.
 workstation processes are applied in the third and fourth verification standard and the simulation run
model, the cycle time for those two models are fully characterization have been determined to fit the non-
standardized with almost no variability. terminating system behavior characteristic. Warm-up
time, simulation length, data collection cycle, and
starting method during replication are the chosen
parameters for simulation run characterization. All four
simulation models are successfully simulated in 20
working days runs with 5 working days as warm up
period. The comparison of system performance
indicators for each model successfully describe the
behavior and characteristic of each model, and give more
comprehensive understanding among simulation model
and current model. The proposed solutions are expected
Figure 9 : Comparison of product cycle time to give benefits to the company in implementing new
design of layout or strategy specifically the truck
assembly line.
4.3 Manufacturing Line Efficiency
6. References
The third performance indicator is manufacturing
[1]. J. Banks, J. S. Carson, B. R. Nelson and D. M.
line efficiency that describes the total system utility. The
Nicol, Discrete Event System Simulation, New
higher line efficiency means that the manufacturing
Jersey, Prentice Hall, 2001.
system is more efficient and utilized, at the same time
[2]. A. M. Law and D. W. Kelton, Simulation Modeling
also means less idle time. Figure 10 illustrated the
and Analysis, Boston: McGraw-Hill Higher
manufacturing line efficiency between current model and
Education, 2000.
each simulation model. Figure 10 likely explained that
[3]. R. E. Shannon, Systems Simulation–The Art and
first model that apply line balancing gave the most
Science. Prentice-Hall, 1975.
significant line efficiency improvement from 61.5 % in
[4]. J. S. Carson, Introduction to Modeling and
the current model until 84.55% in second model (37.5 %
Simulation, Proceeding of 2004 Winter Simulation
improvement).
Conference, 9-16, 2004.
[5]. C. Roser, M. Nakano and M. Tanaka, Throughput
analysis using a single simulation. Nagakute, Aichi.
Toyota Central Research and Development
Department, Proceeding of 2002 Winter
Simulation Conference, 1087-1094, 2002.
[6]. V. Patel, I. Ashby and J. Ma, Discrete event
simulation in automotive final process: Michigan.
General Motors Proceeding of 2002 Winter
Simulation Conferenc, 1030-1034, 2002.
[7]. E. J. Williams and D. E. Orlando, Simulation
applied to final engine drop assembly. Michigan.
Ford Motor Company, Proceeding of 1988 Winter
Simulation Conference, 934-949, 1998.
[8]. Y. H. Park, J. E. Matson, and D. M. Miller,
Figure 10: Comparison of manufacturing line efficiency Simulation and analysis of the Mercedez-Benz
Production Facility, Proceeding of 1998 Winter
Simulation Conference, 132-137, 1998.
5. Conclusion [9]. R. Shady, G. Spike and B. Armstrong, Simulation
of new product workcell, Proceeding of 1997
The solutions is designed based on four approaches Winter Simulation Conference, 739-743, 1997.
and principles that cater how to minimize imbalance [10]. M. P. Groover, Automation, Production Systems,
workloads in assembly line, to improve material handling and Computer-Integrated Manufacturing, New
capabilities through facilities re-layout, and to automate Jersey, Prentice Hall International, Inc., 2001.
the processes. The initial model has been developed in [11]. J. A. Tompkins, J.A. White, Y.A. Bozer, and
term of process logic, animation and interfaces using J.M.A. Tanchoco, Facilities Planning. New Jersey:
ARENA version 7.1 Simulation application package. John Wiley and Sons, Inc., 2003
The model meet all requirement of ARENA model

786

Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on January 11, 2009 at 20:11 from IEEE Xplore. Restrictions apply.
 Figure 2: Production Process Flow

Figure 4: Line Balancing Strategy (First model) Figure 6: Separated MHS (Second model)

Figure 5: Parallel Operation Strategy (Third model) Figure 7: Full Synchronized MHS (Fourth model)

787

Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on January 11, 2009 at 20:11 from IEEE Xplore. Restrictions apply.