CHAPTER 2

LITERATURE REVIEW

2.1 INTRODUCTION

In any research study, the work done in the past forms the
foundation of the work to be carried out in future. It is very much important to
understand the past work for two major reasons, first the past work gives the
trend and pattern followed in the past and next it helps to avoid duplication of
any work carried out in the past and throwing lot of insights to develop the
current focused area.

The investigation of the past work on automotive body quality and

productivity is identified in five major areas. They are

1) Body dimension variation analysis and methodologies followed

to address various quality issues

2) Various investigations carried out to improve productivity and

quality in body shops

3) Lean manufacturing & optimizations carried to solve

manufacturing issues

4) Investigations carried out with design of experiments

methodology to solve various problems.

5) Resistance spot welding optimization

Certain key research papers which are of greater importance with

possibility of practical applications are taken up for review.

27
2.2 BODY DIMENSIONAL VARIATIONS

While assembly of body in white (BIW), various sub-assemblies &

parts are welded together by holding them with the help of different fixtures.
To achieve the quality, fixtures should be with proper design and geometry.
Accuracy of body in white (BIW) assembly depends on the accuracy of fixture
and processes.

Body assembly sheet metal dimensional variations due to various

welding sequences was studied and the author has recommended few
design guide lines to the sheet metal designers to select weld sequences to
set appropriate geometry, design guide line provided to make stress free
welding and stress relief patterns for different joineries [28]. The dimensional
variation effects of positional variation of spot weld was studied by Mali et al
and he has concluded that; “Spot weld position variation control is essential
to maintain proper assemble dimensional accuracy” [29]. Liu et al has studied
the panel surface resistance and observed that the quality of spot weld varies
with change in surface contact. Thus, quantity of surface contacts and
magnitude of surface contact play an important role in the weld and
dimensional integrity of the BIW [30]. Doshi has done a case study based
research to implement the statistical process control method in different
suppliers and observed difficulties due to process variations, also identified
the root causes for process variations [31]. He has tried to resolve the
process variations by implementing various action plans at supplier locations
that also indicates the possibility of process variations in an automobile
manufacturing facility. Johan has studied weld spot distribution among the
robotic welding stations and he has mentioned that “the spot distribution
between the robotic stations and sequence of spots plays a significant role in
dimensional quality and throughput [32].”

A computer-aided intelligent system was developed by Guanlong

et al, which will automatically generate the optimal joint types and assembly
sequences for the best dimensional quality for any given assembly. This is an

28
aid tool for designer to develop assembly joineries based on the existing
library, a best suitable configuration can be obtained from the system [33].

Optimization of welding process and how it leads to the quality in

manufacturing was studied by Hardikar, He has taken an overview of various
techniques used to achieve minimum distortion during the metal parts are
being assembled and concluded that “Reducing dimensional variation is of
critical importance to improve the final product quality”. Software simulation
helps to improve the accuracy of the product in terms of cost, quality and
dimension [34]. Complications of shimming in multi model lines to control
body manufacturing quality was studied by Carsten and recommended force
controlled shimming method to overcome the manufacturing dimensional
quality setting issue in multi model lines [35].

Every spot has its own position as constraint for the dimensional
quality and each sub assembly dimension decides the quality of the final
assembly. kristina Warmefjord et al considered that, there are many factors
affecting the final geometrical outcome of a BIW, spot welding sequence is
one of the identified factor considered to be critical. It is better to choose a
welding sequence that can optimize the variations and deviations of the final
assembly. The author has investigated about the correlation between offset
and standard deviation for different welding sequences. She has presented
the investigation report of number of spot welds required to lock the geometry
with industrial case studies [36].

Geometric variation of an assembly affects the functional points as

well as the quality of the BIW. Rikard soderberg et al described about the
virtual geometric assurance tool for verification of the geometric
characteristics using variation simulation model. He has discussed about that
all manufacturing processes are afflicted by variation which implies that the
nominal value of a manufacturing dimension may not be expected at all the
times. The manufacturing dimension may rather be described by an expected
range and a probability distribution [37].

29
 The diagnosis of dimensional variation can be controlled through
root cause analysis. The resulted variation of an assembly depends on
component variation and locating tool, hence a methodology was developed
to diagnose the multilevel hierarchical assembly variation by Johan [38]. In
the manufacturing process, pre-controlled methods are essential for process
improvements as well as the body quality. Sanches Jr et al described about
the sub assembly pre-controlled methods like master sample preparation,
assembly fixture calibration, process validation and gauge inspection for the
improvement of BIW dimensional stability [39].

Kang et al elaborated about how dimensional variation propagates

through manufacturing process. They have created a new methodology
based on finite element analysis by using parametric model. It is used to
incorporate the input variation from different variables [40].

2.3 PRODUCTIVITY AND QUALITY IMPROVEMENTS

Geometry setting in an automotive body shop is a critical process

and requires an adequate method to plan and execute any new model
introduction. Productivity enhancement is one of the key focused areas in the
current trend of automobile manufacturing. Maximum benefits must be
obtained in every aspect with a minimal environment foot print, Minhaj et al
[41]. A manufacturer faces unpredictable changes in the market and the
market demands a lot of product launches [42]. Jadhav has given a roadmap
for lean principles implementation in the Indian automotive component
manufacturing industry. In this study, the author observes that the Indian
industries need to make a paradigm shift in their strategies and approaches
to improve production capabilities, productivity, quality and scalability [43].

Amir has studied, Hours-Per-Vehicle (HPV) reduction for North

American automobile Industry and reduced around 54.75 minutes on
average each year to improve productivity [44]. Selvaraj et al studied the
simplification of sheet metal components design and manufacturing
processes. They have discussed the ways to reduce the number of sheet

30
metal components through parts integration [45]. Mahmoud et al studied
productivity improvement and resource optimization of an automobile
manufacturing shop [46].

In the competitive automotive market condition, reduction of

production cost is very essential to meet customer demands [47]. Indian
manufacturing industries are focusing to deliver low cost, improved quality
and diverse products with superior performance, Kanwarpreet et al [48]. Al-
Zaher studied and recommends the implementation of the strategy of low
price with high quality factors. The manufacturing ecosystem is one of the
areas which directly affect the control over capital and revenue expenses in a
new model introduction [49]. To control initial investment and to lower the
operational cost; optimization of equipment and process is essential in body
shop process planning. Reduction of geometry stations in a body shop
directly contributes to reduction in capital investment.

Cycle time balancing between body shop welding stations were

carried out by Ali with a multi model objective queuing method. From the
developed model, two results were obtained. First, the process time
balancing between the stations resulted in increased production quantity and
saved 7.81 minutes per day. Second, the reduction in transfer lead-time
resulted in reduction of 5.81 minutes per day, also the author has mentioned
that the lead time reduction was not focused in the previous investigations
[50].

Robot weld load balancing for multi model line was studied by
Johan and proposed a novel generalized simulation-based method for
automatic robot line balancing for robot positioning [51]. Impact of different
cycle times and buffer sizes of the body shop was simulated and validated
with a case study by Sven et al and explained that the maximum utilization of
available cycle time with the possible reduction of production equipment [52].

Jianjun shi et al proposed a process navigator for quality control

and assurance system. It includes the real-time data collection, online

31
tracking and fault identification, inspection data base, variation pattern
animation and knowledge based diagnosis. He provided an effective
measurement tool for productivity as well as quality improvement [53].

DonHee Lee et al empirically studied the effects of manufacturing

systems by using data from the survey of global manufacturing research
group. They have described that “the operational performance can be
improved through best quality management practices and competitive
priorities”. They have tested the research data using structural equation
modelling for hypotheses from various multinational firm from different
countries [54].

Modern revolution has demanded the quality and productivity at a

similar rate. Vinod S. Gorantiwar et al described the importance of various
Critical Success Factors (CSF) and its impact on total quality management
(TQM). The integration of various critical success factors can help the
manufacturing industries in achieving sustainable quality improvement with
the TQM approach [55].

Flexible manufacturing is the boon for automobile industry, quality

management and just-in-time makes the better products to the customer on
time. Chi Anh Phan et al done an empirical analysis by ANOVA and
hierarchical regression from survey data of various manufacturing plants.
They have suggested that the manufacturing plants should highly focus on
Quality Management (QM) and Just in Time (JIT) to explore the linkage and
synergy of operational practices [56].

2.4 LEAN MANUFACTURING & OPTIMIZATION

Optimization is one of the major focus areas to meet the business

requirements. There are some optimization initiatives developed and
deployed in automobile industries; some of those process optimization
solutions are discussed below. Assembly line balancing problem was studied
by Mohd and it was quoted in his thesis that “Research works focus on the
integration of manufacturing optimization between the process planning
32
stage and manufacturing process stage remains limited” [57]. It is evident
from his quote that the scope for improvement is very wide in the integration
of process planning and the execution.

To maximize the production, Chuan has developed an efficient

buffer design algorithm subject to a production rate constraint and provided
valuable insight about production line design to manufacturing system
practitioners [58].

Different research activities were carried out with various

approaches to optimize the processes. BIW geometry setting and process
optimization is a very wide subject, where there is a huge scope for
improving the unresolved issues and partially addressed issues. Variation
simulation for optimization of manufacturing process is essential for quality
parameter establishment [59].

Rakesh has discussed the need of lean manufacturing

implementation to improve the operational parameters such as productivity,
quality and cost. Manufacturing process, layout, supplier, training of work
force and scheduling are considered as main factors in the study. The results
were concluded with that, the plants which uses Quality Management and
Just in Time techniques achieves high performance [60]. Quality control cost
optimization issue was studied by Jian and recommended to decrease the
rate of false alarm to improve the economic condition in body manufacturing
[61].

Jaiprakash Bhamu et al described the systematic way of reducing

the wastes in lean manufacturing by value stream mapping. He demonstrates
the effects of value stream mapping through case study and concluded that
lean implementation is a tool for improving productivity and quality of a
company [62].

As stated by Ashish gupta et al, during process planning stage in

an automotive vehicle development process (VDP), one of the tasks is to
select tools such as weld guns and fixture from large tool library and assign
33
them to individual manufacturing operation. Software packages merely pick
and validate individual library items to find a solution. Further, no design
assistance is provided to configure a valid new tool. Currently in automotive
industry, the tool selection process is manual, local & subjective. It leads to
proliferation of tool variants resulting in increased costs and reduced
operational flexibility [63]. In his paper, he has proposed an approach for
automated and optimized method of selection of tools.

According to Osma m Erfan et al, the complete elimination of

waste is the target of any qualified system. This concept is vitally important
today. Since, in today’s highly competitive world there is nothing we can
waste. Value stream mapping tool was used to expose the waste and identify
a proposed plan for improvement. This results in the system of being more
productive, flexible, and smooth and with high quality service [64]. From his
analysis it is clearly observed that elimination of waste such as excess
inventory in the system is very critical.

According to Endris et al, Lean Product and Process Development

(Lean PPD) is an emerging paradigm wherein lean thinking is considered in
the design and development of engineering products. Although lean design
and lean product development have received attention in previous works,
process development has received little; research in this area is necessary to
improve the concurrency of product and process development. Integration of
manufacturing knowledge at the conceptual design stage is of paramount
importance to increase the designer’s awareness of manufacturing
capabilities and opportunities. As a result, design rework can also be
minimized and possible manufacturing plans can be effectively explored [65].

Arnout Pool, et al has concluded through various case studies that

it is impossible to create a theory to describe or predict the leaned system &
it is a continuous activity [66]. Karlsson et al suggested a check list for what
to aim at when trying to implement lean production. In his study he has
concluded that lean should be a direction, rather than as a state to be

34
reached after a certain time and, therefore, the focus should be on the
changes in the determinants, not on their actual values [67].

Yu Cheng Wong et al has studied the adoption of lean

manufacturing in the electrical and electronics industry. They have
considered 14 key areas of lean manufacturing in their study, scheduling,
inventory, material handling, equipment, work processes, quality, employees,
layout, suppliers, customers, safety and ergonomics, product design,
management and culture, and tools and techniques [68].

Rose et al has studied the lean implementation in Malaysian auto

components industries to understand the level of perception and
implementation of lean manufacturing. Most of the surveyed companies
agreed on the importance of lean practices in their companies, but the level
of implementation was not perceived by them in the same ways. To improve
this condition, implementation of lean manufacturing across all the industries
is considered critical [69].

Om Prakash Yadav, et al has investigated the level of lean

implementation in the global automotive industries. Also, the study explains
about the various learnings from the actual implementation practices
particularly in the USA, UK and Indian automotive sectors [70].

2.5 DESIGN OF EXPERIMENTS

The impact of an input variable on a process can only be quantified

by a series of experiments that are performed with varying levels of the input.
Ricardo B P et al observes that the trial and error procedure involved in the
conventional process of experimentation can be avoided by structuring the
flow of experiments using Design of experiments (DOE) [71].

The objective of the Taguchi method is to measure the sensitivity

of the output to the variation of input parameters. Thus, it is critical to choose
only the parameters that have the highest impact on the outcome, figure 2.1
[72]. Spot quantity, welding surface contact, spot sequence and spot position
35
were selected for the study due to the significance of the parameters on the
output in the study presented in this thesis.

Rao et al observed that, in the selection of the appropriate

orthogonal array for DOE, the total number of experiments should be equal
or higher than the total number of levels of all the parameters. This ensures
equal occurrence of every parameter combination. Hence, for this study, an
L-27 Taguchi orthogonal array is selected [73].

PARAMETER IDENTIFICATION

PARAMETER LEVEL ESTABLISHMENT

TAGUCHI ORTHOGONAL ARRAY IDENTIFICATION

DESIGN OF EXPERIMENTS

EXPERIMENTATION

RESULTS & ANALYSIS

Figure 2.1 Flow chart of Design of Experiments

Niranjan Kumar Singh et al presents a systematic approach to

determine the optimization and effect of weld process parameters by Taguchi
method. He described the optimum weld parameters determined by L32
Orthogonal array experimental method. ANOVA and F-test is selected for
determining the significant parameters affecting the spot weld parameters
[74].

36
 Optimization of a process requires evaluation of the effects of
various process parameters on the output. This entails sequential
experimentation using various input parameter combinations. Taguchi
method which was developed by Genichi Taguchi, Mikovik et al provides an
experimentation process which can achieve the objective of process
optimization with minimum iterations. The goal of this method is to use the
output of the experiments to find the Signal: Noise (S/N) ratio of the input
parameters. The S/N ratio shows the quantum of the impact the input has on
the process [75].

The S/N ratio can be measured in three categories; “Larger is

better,” “Nominal is better” and “Smaller is better.” An appropriate category
should be selected based on the requirement, Fazeli et al [76].

2.6 RESISTANCE SPOT WELDING OPTIMIZATION

Resistance spot welding is widely used for BIW manufacturing all

over the globe. The recent research trend has focused for optimization of spot
welds for a better quality of the sheet metal assemblies. Hsien-Yu Tseng
proposed a solution for welding economic design problem. He applied the
general regression neural network for economic welding design construction
and can select the optimum weld parameters based on genetic algorithms [77].

Soo-Won Chae et al developed an optimal design system using

adaptive meshes for spot weld process optimization. He has proposed an h-
version background meshes for optimization of spot welding process [78].

Coriteac Let al showed that spot welding optimization can be done

with algorithms by finite element models. In his case study, the optimization
factors were explained for increasing production volume and cost reduction
[79]. The strength of the spot weld depends on the weld parameters (weld
pressure, current, weld time). Shruti Naik et al explained the rupture of spot
welds and suggested that aluminum material is preferred for spot welding,
but it requires much higher tip forces than steel [80]. Yong Xia et al have
analyzed the concept of design for manufacturing. The optimization in spot
37
welds, results that the structural torsion stiffness is relatively sensitive to the
layout of spot weld. He has done a foundation research towards the
structural performance assessments and the weld layout design [81].

The optimization and performance evaluation tools are inevitable

for manufacturing processes. Tempelmeier developed an algorithm system
software, which is helpful for the planners to plan manufacturing process in
less time [82]. Heui-Bom Lee et al done the distortion analysis in spot welded
area of Body-In-White. The distortion from welding sequence can be
minimized, if clamping system is perfect. The weld force originating from the
assembly tolerance is a critical design factor that determines the weld
sequence and the clamping system for minimum distortion. The durability of
BIW is mainly controlled by the strength of the spot weld joints [83]. Andreas
Rupp et al done a theoretical approach for structural durability and the
reliable design for optimization of spot welded components. Finite-Element
post processor calculation procedure allows to determine the local stress and
improve the life of the BIW structure [84].

Chon L.Tsai et al studied the control systems for the resistance

spot welding. They have done a systematic approach for expansion-based
control algorithm for spot welding, the inadequate welding conditions that
causes nugget expulsion and current shunting [85].

Xin Sun et al addresses the modeling issues of resistance spot welds.

He explained the techniques on simulation of weld process, weld property
prediction and evaluation of weld engineering performance [86]. Won Wook
Jung et al discussed about the accurate reason for the fatigue failure of
the spot weld in the developing stage and another method for repair [87]. Gilvan
Prada Rossi presented to compare two different methods of spot weld modeling
through a correlation study, by considering the maximum force of the welded
joint, between virtual and lab test results [88].

Shinichiro Yoda et al described a new method to predict the

rupture of spot welds, which is suitable for vehicle crash simulation. He had

38
developed a new method to predict the rupture of spot welds which relates
axial and shear forces and bending moment of spot weld that can cause
stress around nugget. Rupture risk can also be estimated through this
method [89].

Eric R. Pickering et al studied aluminum spot welding.

When aluminum sheets are spot welded, electrode tip life is limited due to tip
erosion and pickup of aluminum on the tip. They have examined the control
variables in the spot welding process and offered an improved weld schedule
to achieve desired weld properties [90].

The geometric integrity of a sheet metal assembly is a direct function

of the spot weld positions and the spot weld count. The geometry of a sheet
metal assembly can be controlled by a minimum of two spot welds and a
maximum of three spot welds depending on the part parameters, Selvam G et al
[91]. When an assembly is controlled by such a low count of spot welds, the
quality of the spot welds is paramount to keep the geometric integrity intact.

Failure resistance of a spot weld can be improved by optimization

of weld parameters such as weld current, weld time and electrode geometry,
Nasir Z et al [92]. An automotive structure has a wide variety of combinations
which require the process planner to tune the weld parameters for every weld
station individually. The parameter setting, and weld coupon evaluation
consumes a high process time. Commonization of parameters will provide a
substantial improvement in this area [93]. Selvam G et al observes that one
of the crucial aspects that impact the total automotive Body-in-White (BIW)
proto build timeline is the parameter setting and fine-tuning time [94].

The tensile-shear strength of the weld is dependent on the

thickness of the sheet combinations and the diameter of the nugget formed,
Pouranvari M et al [95]. The change in the weld parameters will impact the
developed weld nugget size, Shamsul J.B et al. [96]. Thus, by ensuring a
target range of nugget diameter for a thickness combination, the weld
parameters can be studied for optimization.

39
 Different material grades can affect the spot weld quality and
strength, Gedgeon S.A et al. [97]. As an automotive BIW is made of multiple
material combinations, optimization of weld parameters will provide effective
results only when the commonization of parameters is across multiple
material combinations. Thakur A.G et al studied the impact of weld
parameters on spot weld quality of galvanized steel. The weld parameters;
weld current, weld time, electrode diameter and weld force were optimized
using Taguchi experimentation process [98].

2.7 GAPS IDENTIFIED IN CONVENTIONAL PROCESS PLANNING

The existing data sources indicate that there is no attention

shown in appropriate spot allocation between the geometry and
re-spot stations

Percentage of spots and stations are having an inter-linkage.

Reduction of geometry spots will reduce the number of
geometry stations

There is a scope to bridge the identified gap in BIW process

planning to improve productivity

To the best of our knowledge, no previous work has been reported

on “Appropriate spot allocation between the geometry and re-spot stations to
improve productivity”.

And, it was noticed that less attention was given on “Improving

productivity in body shop with geometry spots optimization and reduction of
geometry stations” and hence this work has been carried out to bridge the
gap.

2.8 SUMMARY

From the literature review, it is clearly observed that the spot weld
accuracy, position of spot welds, sequencing of spot welding and the number
of spots play an important role in body dimensional control. Different
40
research activities were carried out with various approaches to improve the
dimensional accuracy and to optimize the processes. BIW geometry setting
and process optimization is a very wide subject, where there are huge
scopes for improving the unresolved issues and partially addressed issues.

In the latest research of Kristina et al, it has been identified that two
weld spots are sufficient to set the geometry between any two assembled
parts in any sub-assembly welding [36]. This was taken as the basis and
further studies were conducted with the parts of different sizes and with
different surface contacts.

Ali explained about the importance of productivity improvement in

body shop by reducing transfer lead-time and process time. However, BIW
geometry set and process optimization are a vast area and still lot of un-
answered questions are there in front of us to resolve [50].

In this study both geometry set, and process optimization are

integrated to obtain maximum productivity with an appropriate dimensional
quality. As per the previous studies, the minimum number of spots required
to set geometry between the two parts are “2”. Further investigation is done
with the consideration of four major factors to investigate the applicability of
implementing the optimized number of spots across body shops with different
configurations. This problem is studied using Design of Experiments (DOE)
method. Abbas and Kawther used Taguchi (L27OA) method for tableting
process optimization with three controllable factors [99], there are four
controllable factors used in our study.

41