1

CHAPTER 1
Introduction
Six Sigma management is the philosophy of the reduction of variation in all critical
process to achieve continuous and breakthrough improvements. Lean manufacturing
seeks to provide optimal quality by eliminating waste. Lean Six Sigma is the combination
of both. Kwak and Anbari (2006) have defined Six sigma as a method of project driven
management approach to improve the organizations products, service, and process by
continually reducing defects in the organization. Lean manufacturing and Six Sigma are
powerful philosophies baked by several tools for improving quality, productivity,
profitability and market competitiveness for any corporation in a holistic manner
(Cudney, Mehta and Monroe, 2006). In the competitive environment, the fundamental
goals of the company are to be able to survive in the market and for the long term.
Productivity should be evaluated as one of the most important indicators of the business
performance. Product quality is now measured by the level of error in the millions. To
achieve these strict quality criteria, the whole system starting from the design step should
be constructed such that it can produce right at first time. From the last 15 years many
leader companies implementing lean six sigma methods. Lean Six Sigma is an integration
of Six Sigma and Lean Manufacturing. Six Sigma provides an integrated improvement
approach that increases quality by reducing variation, defects, and costs. Six sigma
focuses on quality rather than speed, where speed up can be resolved by Lean
management. Lean management is better for the improvement in the speed and process
flow rather than in improvement quality. Therefore, the best result can be achieved by
combining both. Lean Management focuses on eliminating loss in process and reducing
the complexity.
Lean management can be implemented in all the fields and bring out opportunities to
increase performance. Six Sigma provides quality philosophy and is a statistical tool to
monitor process performance. It aims to reduce the variability in the process and to
eliminate errors. Lean management and six sigma work together successfully.Six sigma is
applied to improve the quality of the product and Lean management is to make cost
effective and faster. Lean Six Sigma is a combination of certain tools and techniques to
provide Six Sigma practitioners another philosophy to reduce process and production
times while minimizing the variation and reducing waste at the same time (Khalil, Khan
and Mahmood 2006). Six sigma is a philosophy and deployed by carrying out
improvement projects. Lean adds tools that increases process throughput by eliminating
waste. The primary focus of Lean is on reducing waste, synchronizing flows and
managing variability in (process) flows. Lean and Six Sigma have complementary
benefits. For integration, Lean may use the management structures that Six Sigma offers:
Six Sigmas project-by-project approach provides an effective embedding framework to
apply Lean principles.
2
1.1Motivation
In the modern competitive market product performance is being measured in terms of
defect level per million opportunities. From the study of different papers and books now
it is known that in six sigma quality control the number of non-conformities per million
opportunities is only 3.4 when mean is not change. It is very few amounts as compared
with traditional system.But in Bangladesh the use of Lean Six sigma yet have not
considerable. Now-a days Lean tools is being used in some manufacturing industries. For
this reason we have been motivated to work on Lean Six sigma and applied in any
manufacturing company in Bangladesh.
1.2Research Objectives
The objectives of this work is to ensure high level of product performance in terms of
defect level and also to reduce the cost associated with the production, improvement of
the process quality and enhancement of the companies goodwill. To achieve these
qualities Lean and Six sigma have been used in the form of integration. The use of six
sigma methodology is to reduce defect per million opportunities and to reduce process
variation sigma () and Lean tools to reduce costs, waste and speed up the process.
1.3Problem statement
It is known to all that, there are seven wastes in any type of manufacturing industries.
These are overproduction, waiting, transportation, inventory, motion, over-processing and
defective items. The purposes of this research papers is to reduce mainly two among the
seven wastes and thats over-processing and defective items using Lean and Six sigma. If
it is possible to reduce the process variability () by improving process of production,
then number of defective items will be also reduced. Lean is a tool to reduce waste and
Six sigma is a philosophy. If the implementation of Lean Six sigma be successful then the
number of defect and reworks will be reduced. But in Bangladesh the implementation of
Lean Six sigma yet have not considerable. Most of the manufacturing industries are not
concerned about Lean Six sigma. So that, there is big amount of loss due to defective item
and rework. It has been seen that in Pran Agro Ltd. one of the renown food product
industry in Bangladesh. In the ice-pop department in this company it has been found that
there are mainly five types of defects. These are leakage, black particle, without coding,
excess/short, cap loose. There is big amount of losses due to these defective items. The
main purpose of this thesis paper is to reduce these defective items by using Lean Six
Sigma tools. If this can be reduced then a surplus profit can be achieved. And it is
possible by successfully implementation of Lean Six sigma methodology.
3
1.4 Methodology
This section presents the methodological framework that will be used for the subsequent
empirical analysis. After conducting a literature review, the core of this research is data
collection. Prior to visit and a detailed questionnaire is developed with the aim of finding
out how many defect and rework are being happened regularly and what caution they
follow to reduce them. The research will primarily intend to be done in two stages: an on
field questionnaire survey and upon response and face to face interviews will be
conducted to get closer insights and their corresponding response. Secondary material
will also be used, literatures, research publications, trade and production data, articles in
local news paper and internet. The study will also be included some informal
communications and direct observation which will give added depth. After the data
collection we have to analysis the problem related to the defective item. Calculation will
be according to the DMAIC approach. In different Phase of DMAIC model analysis will
be done by different types lean tools and Six sigma tools, such as control chart, Fish-
bone diagram, Pareto analysis and QFD. After observing the controlling situation in
control chart and finding out the causes of the defects by using cause effect diagram, it
have been tried to improve the condition by implementing Lean tools(5s).
1.5 Organization of thethesis
The structure of this Thesis paper has been organized in following manners. In the first
chapter of the thesis contain Introduction with subtitles motivation, research objectives,
problem definitions, methodology and organization of thesis. Literature review in the
second chapter contains some related past Thesis work on Lean manufacturing and Six
Sigma and how different this paper from them. A theoretical framework of the thesis has
explained in chapter 3.All the tools and theory that have been used in this works has
explained briefly. The case study and data collection is discussed with data about the
concerning company in chapter 4. After that, in chapter 5, all the calculations are showed
with chapter named Result. Result analysis has been conducted in chapter 6. Finally some
recommendations and future works are given in chapter 7. This paper has been completed
through conclusion, reference and appendix successively.
4
CHAPTER 2
Literature Review
Six sigma has been implemented by Motorola in the early in 1980s and Lean
manufacturing has emerged towards the end of the 1990s. It has become so popular
methods and become prevalent in all business areas over time. There are lot of work on
six sigma and Lean management and combination of both. For the last one decades
history in the literature, a number of academic terms has found about this methodology.
ONeil and Duvall (2003) in their worked they have focused on post occupancy
evaluation (POE) research method and discussed about the application in six sigma
quality frameworks. They have used POE to create and manage the optimal space for the
office workers. The team began to process of tracking these data using Minitab, which
contain six sigma quality tools- such as control chart. Does and Koning (2006) explained
that the use of lean six sigma in a hospital to provide the healthcare to the patient at
lowest possible cost and time. They solved the most important problem of their patients
(long waiting time) and cost as well. They had tried to reduce the lead time which is
perhaps one of the most important quality indicators from the perspective of their patients
in healthcare. How Lean Six sigma can be applied in non-manufacturing organization is
their future work. Kwak and Anbari (2006) have identified the benefits, obstacles and
future of six sigma approach. They focused on examination of the evaluation, benefits
and challenges of six sigma practices and finding outs the key factors which leads the
success of six sigma. Factors influencing successful six sigma projects include
management involvement and organizational commitment, project management and
control skills, cultural change, and continuous training. However, integrating the data
driven, structured six sigma processes in organizations still has room for improvement
which is further research objective.
In the recent years, the manufacturing industry has successfully applied the six sigma
methodology to project. However, due to insufficient data or a misunderstanding of the
six sigma methodology, some of the project failed. Aimed to discuss define, measure,
analyze, improve and control (DMAIC) phases of integrating a manufacturing execution
system (MES) and six sigma methodology (Hawang 2006). Chang and Wang (2007)
explained six sigma methodology is a method that can lead to a continuous decrease in
process variance. In their work they applied six sigma methodologies and proposed a
continuous improvement model on different phase of collaborative planning, together
with forecasting and replenishment (CPFR). Future work should pull data from longer
period of time and investigate how to manage demand smoothing and supply arrangement
in CPFR. Fairbanks et al. (2007) discussed about the improving patient flow in the
preoperative environment is challenging, but it has positive implications for both staff
members and for the facility. Improve patient throughput by incorporating six sigma and
lean methodologies for patients undergoing elective procedures.
Ditahardiyaniet.al (2008) has presented the six sigma methodology and its
implementation in a primer packaging process of Cranberry drink. DMAIC approaches
have used to analyze and to improve the primer packaging process, which have high
variability and defects output. Hekmatpanah et al (2008) in their works they has surveyed
the six sigma process and its impact on the organizational productivity. And they have
studied key concepts, problem solving process of six sigma as well as the survey of
important fields such as; DMAIC, six sigma and productivity applied program, and other
advantage of six sigma. Amer et al. (2008) in their works they focused mainly on the
5
customer requirement and controlling supply chain variables. By optimizing supply chain
processes and meeting customers requirement order can be fulfilled. Here fuzzy set
theory provides a way of monitoring supply chain performance. They recommended that
further work should address the strategic outcomes of improved collaborative
performance that the methodology could potentially deliver.
Chakravorty (2009) provided a model to effectively guide the implementation of six
sigma is to reduce variation or waste from the operation. The purpose of this research was
to develop an effective implementation model which consists of six steps. The first step is
to perform strategic analysis driven by the market and customer. The second step is to
establish a high level, cross-functional team to drive the improvement initiative. The third
step is to identify overall improvement tool. The fourth step is to perform high-level
process mapping and to prioritize improvement opportunities. The fifth is to develop a
detailed plan for low-level improvement teams, and the sixth step is to implement,
document, and revised as needed.Yang and Hsieh (2009) suggested that six sigma is a
tactical tool of significant value in achieving operational excellence. The project selection
decision, under a resources constraint, is the early stage of implementation for a six sigma
intervention. They also proposed to adopt national quality award criteria as the six sigma
project selection criteria and proposed a hierarchical criteria evaluation process.Schon,
Bergquist and Kiefsjo (2010) presented a study of how six sigma influences job
satisfaction among employees at three large companies with manufacturing units in
Sweden that have used different implementation strategies.Barac, Milovanovic and
Andjelkovic (2010) have applied Lean six sigma methodologies in supply chain
management in manufacturing products. They have tried to eliminate non-value added
process and waste in terms of time, cost or inventory. Waste reduction and the removal of
unnecessary process can save companies millionsof dollars a year. Getting the right
product at the right price, at the right time to the endcustomer is not only key to the
success of companies in competitive markets, but also thekey to their survival.
Six sigma focuses on quality more than speed. Lean management removes the weakness
of six sigma by speeding process. The Lean six sigma methodology developed by using
these two techniques together is presented. The goal of application is making process
Lean and increasing the Level of sigma. Implementation studies can be carried out in
service and public sectors where Lean six sigma practices are rather inconsiderable is
their future research objective (Atmaca and Girenes, 2011). Padhy and Sahu (2011)
proposed two-stage methodology to identify and select of six sigma project based on (1)
Real Option Analysis for evaluating the value of the project to improve the managerial
flexibility (2) a zero-one integer programming model for selecting and scheduling an
optimal project portfolio, based on the organizations objectives and constraints. His
further research can be carried out in the area for quantifying the impact of the project and
organizational specific risks on project payoffs, which is very much prevalent in the six
sigma projects. The reliability estimate of project cash flow and investment calculation is
another research area. Brun A. (2011) explained the result of a research project focused
on six sigma implementation processes, with a particular attention to understand which
the situation of the enterprises operating in Italy is and consequently, which are the
managerial implications of a six sigma implementation in the typical Italian company. He
also said that any six sigma implementation aims at improving customer satisfaction, by
mean of improved process capability. This, in turn, is made possible by focusing on
Critical to Quality (CrQ) characteristics and implementing improvement action seeking
to continuously reduce process variability in terms of CrQ. These actions are carried out
6
by involving every employee. His further research objective was validation and
Reorganizing it could constitute a road map for implementation of six sigma.Johannsen,
Leist and Zellner (2011)in their work aimed to systematically identify key problems of
the six sigma application in service and to assign them to the phases to be able to
determine the exact moment of their beginning. The lacking process orientationwithin the
framework of the definition of key performance indicator wasthe most important
problem. The second most important problem was seen in theinsufficient data quality in
services which makes the design of adequatemeasurement systems difficult. Their further
research have recommended that about answering the following questions a) What are the
effects of a missing process orientation for a Six Sigma project in services?b) How can
the degree of process orientation be measured in an organization and which degree is
sufficient for a Six Sigma project?
DMAIC (define measure, analyze, improve and control) method in six sigma is often
described as an approach for problem solving. DMAIC is applicable to empirical
problems ranging from well-structured to semi-structured, but not to ill-structured
problems or pluralistic messes of subjective problems. Six sigma is generic method. The
advantage of such methods is that they are very versatile. The disadvantage is that task-
domain specific method can be more powerful because they can be more specific and
operational in the guidance they can provide. Mast and Lokkerbol (2012) has highlighted
the characteristics of the DMAIC approach and its limitation; specifically from problem
solving perspectives. Their future work is to find out direction where the approach may
be improved.
There are lots of papers or works on Lean six sigma methodology in the history of
literature.In this thesis work lean six sigma methodology have been implemented through
DMAIC model in a manufacturing industry to reduce waste. The difference of this work
from the others is in terms of tools used in conducting the thesis and its perspectives.
Here Lean tools have used to reduce the defective items and reworks and Six sigma tools
like control chart, fish-bone diagram, Pareto analysis etc are used to analyze and
determine what is to be controlled. In the previous works these type tools have not used to
reduce waste in terms of defects and reworks in a manufacturing industry in Bangladesh.
7
CHAPTER 3
Theoretical Framework
In this chapter has discussed all the theories and tools that have been used in this thesis
work. This work has been carried out on two philosophies and their related tools. One is
Six sigma management and another is Lean manufacturing. All the TQM tools can be
used as a Six Sigma tools. Some of their have been used in this work likes control chart,
Pareto analysis, fish-bone diagram etc. MCD approach and QFD also has used to identify
the problem. 5S tool of Lean manufacturing have used to recommend the solution. All
tools and philosophies are discussed individually in the below.
3.1 Six Sigma Management
Six Sigma is a short-cut for saying six standard deviations from the mean, which specifies
a tolerable range. The common definition of Six Sigma management is that it is an
organizational initiative designed to create manufacturing, service, and administrative
processes that produce a high rate of sustained improvement in both defect reduction and
cycle time. By using it Motorola has been successful to reduce 50% cycle time and
defects within two years. This company is the first one who gets Malcolm Baldrige
National Quality Award in 1988. The award strives to identify those excellent firms that
are worthy role models for other business. The innovation of Motorola which leads to get
the award that was Six Sigma program. The term Six Sigma is derived from the normal
distribution used in statistics. Many observable phenomena can be graphically
represented as a bell shaped or a normal distribution as illustrated in figure 3.1.
LSL
USL
-6 -3 0 3 6
Figure 3.1: Normal distribution of 6 quality control (Mean , standard deviation )
A statistic is used to measure the typical value of output of a process is called mean or
average, .Another statistic is which is used to measure variability of the output of a
process. In the normal distribution of Six Sigma quality control the interval is calculated
by mean plus or minus 6 standard deviations thats mean 6. This interval contains
99.9999998% of the data. That means 2 data per billion of data values are outside of the
area created by the mean plus or minus 6 standard deviations (2=1,000,000,000x
[0.0000002%=100%-99.9999998%]). Motorola presume that mean can be drift by 1.5 in
8
either direction due to the natural causes. A process which is operating on Six Sigma
quality control system per million non-conformities will be 3.4 if mean shifted by 1.5 in
either direction. In the 3 quality control system the interval is obtained similarly by
mean plus or minus 3 standard deviations that means 3. It contains 99.73% of the
data. This means 2700 defects per million opportunities will be outside the area created
by mean plus or minus 3 standard deviations (2,700=1,000,000 x [0.27%=100%-
99.73%]). If mean is shifted by 1.5 in either direction from the mean than number of
non-conformities per million will be 66,807.
The difference between the 3 quality control management and 6 quality control
management is on VoP (voice f process) and VoC (voice of customer). In 3 quality
control VoP is equal to VoC. But in the 6 quality control system VoP is half of the VoC.
So that the value of in Six Sigma Quality control system is half of the value in the three
sigma quality control system. Six Sigma promotes the idea that the distribution of output
for a stable normally distributed process (Voice of the process) should be designed to take
up no more than half of the tolerance allowed by the specification limits (Voice of the
customer). Although process may be designed to be at their best, practically it has been
seen that the process variation may increase over time. This variation may caused by
small variation in inputs, changing in monitoring system, changing conditions. This
process variation caused the shift of process mean in either direction. It has been seen that
in practice the average shift in mean is about 1.5 standard deviations from target due to
the increase in process variation. In the 1980s, Motorola demonstrated that in practice, a
1.5 standard deviation shift was what was observed as the equivalent increase in process
variation for many processes that were benchmarked. Three Sigma quality control
managements graphically can be shown in normal curve in the following figure 3.2.
LSL USL
Figure 3.2: Normal curve of Three Sigma quality control
In 3 quality management system the upper specification limit is calculated by
USL=+3 and lower specification limit is calculated by LSL=-3. Where is the
mean and is the standard deviation of the process. Voice of customer is equal to the
voice of process in this control management. Large number of product is gone out of
control if mean shift in either direction from the origin.
-3 3
9
Similarly in the Six Sigma quality control system upper specification limit is calculated
by USL=+6 and lower specification limit is calculated by LSL=-6. Here also and
is expressing similar meanings as three sigma quality control management. It is known
that in 6 management VoP is half of the VoC.VoC gives the specification interval
among which customer will accept the product satisfactorily. Outside this specification
limit product is considered as a non-conformity product. Since VoP is half of VoC so that
the process will try to control their product within half of the specification limit, that
means within 3 limits. But it should be remind again that the value of sigma in 6 is
half of the value in the 3 quality control management. The specification limit in both
case is same and given by the customer. That means 3 specification limit in the 3
quality control management and 6 specification limit in the 3 quality control
management contains same limit. The difference between two is on the VoP which is
designed by the company. In the 3 quality control VoP is designed equal to VoC and in
the 6 quality control VoP is designed equal to the half of the VoC. The mean or average
is generally situated in the middle position in the interval. But due to the natural deviation
mean may shift either direction even process is well designed. The graphical presentation
of normal distribution with mean shift in 6 management is shown in figure 3.3.
-1.5 1.5
LSL USL
-6 -4.5 -3 -1.5 0 1.5 3 4.5 6
Figure 3.3: Six Sigma process with 1.5-Sigma shift in the mean
The mean shifted by 1.5 results in a 3.4 non-conformities or defects per million
opportunities at the nearest specification limit, or one or late monthly report in 24,510
years [1/0.0000034/12]. This is the definition of 6- Sigma level of quality. Whereas in the
3- Sigma process if mean shift by 1.5 then results in 66,807 defects per million
opportunities at the nearest specification limits. The difference between the 3-Sigma
process and 6-Sigma process is dramatic enough to certainly believe that 6-Sigma level of
performance matters, especially with more complex processes with a greater number of
steps or components.
10
3.1.1 DMAIC Model
DMAIC model is an acronym for Define, Measure, Analyze, Improve, and Control. It
contains five individual steps through which Six Sigma can be implemented. The
relationship between voice of customer and voice of process and DMAIC models is
explained in figure 3.4. The output distributions of 3-Sigma process and its old flowchart
have been shown in the left side in the figure 3.4. The right side shows a new flowchart
with its 6-Sigma output distribution. The model utilized in Six Sigma management to
move from the old flowchart to the new flowchart through improvement of a process is
called the DMAIC model.
Figure 3.4: Relationship between the VoC, the VoP, and the DMAIC model
The Define phase of a six Sigma DMAIC model is used to identify the product quality
characteristics which is critical to customer (called CTQs). In this thesis paper we have
tried to identify this though QFD model and questionnaires that means voice of customer.
The second phase named measure is involves defining the CTQs operationally.
Start Start
Stop
Ye
s
Stop
No
No
Ye
s
DMAIC
Model
LSL
USL
-3 0 3
OLD
LSL
USL
-6 0 6
NEW
11
The analyze phase involves identifying input and output variables that affect each CTQs
(Xs) using process map or flowchart, creating a cause-effect-diagram to understand the
relationship between the CTQs and Xs. In this phase normally finding out of critical work
or major problem which is responsible to the large part of the problem is identified. The
improve phase deals with the activity related to the improvement of the project. This
phase is involved designing the appropriate experiments to understand the relationships
between the Xs. The final phase of the model named control phase involves avoiding
potential problems in Xs with risk management and mistake proofing, standardizing
successful process changes, controlling the critical Xs, development the process plan and
documentation of the process plan.
3.1.2 Control chart
Control chart is one of the most important, popular and widely used tool of the seventh
basic tool of TQM. This chart shows the output variation of the process graphically. It
contains prefix upper and lower specification limits. In 1924, Walter Shewhart of Bell
Laboratory introduced the concept of control chart as a tool of showing the variation of
the quality characteristics. To manage the process economically and to bring the process
in statistical control system Shewhart had stressed. The variation of the output of the
quality characteristic may caused by several reason or source. The source of variations
can be machine, operators, materials, etc. These types of reason bring a certain amount of
natural variation in the entire production system. The main purpose of control chart is to
show the variation of output and subsequently control it. There are some randomly
occurred causes which are uncontrollable, natural, and result in small variation. These
variations are referred to as a stable system of chance cause.There is another type of
causes of variations which are assignable, with non-random pattern of occurrence. This
type of cause may take place in large amount which may result in the out of control
situation. This type assignable cause may be associated with the machine, method, man,
material etc. assignable cause do not occur frequently if the process control is good and
stable.
The prior signal about the possible out of control situation may also be obtained through
the control chart. So this chart can be acted as a diagnostics of the possible out of control
state of the process. The process which is in control situation is considered as a stable
process and the process which is out of control situation is considered as a unstable
process. When the out of control situation is observed in the control chart then it is up to
the operators, Engineers or management of the process to find out which reason behind
this out of control situation and trying to solve that. At last we can say that the main
purpose of control chart is to reduce the process variability through eliminating the reason
which is for out of control situation. The purposes of Six Sigma quality control
management philosophy is same as control chart so that this chart can be used as an
effective tool of Six Sigma.
A control chart is basically a graphical monitoring tool to observe the pattern of variation.
There are two control limits exist for this purpose:
1.Upper Control Limit (UCL)
2. Lower Control Limit (LCL).
12
A central line also exists between the control limit and it represents the target mean of the
quality characteristics. It is expected that all the data values of the output quality
characteristics will be fallen within this two limits named UCL and LCL. If any data goes
out of this limit then it will be considered as an out of control situation. There need to
make an investigation to find out the reason for which that value is out of the control limit
and after that need to solve it. To draw a control chart the data value of quality
characteristics have to be taken and should have to plot in the control chart. A typical
control chat shown in the figure 3.5.
Rejection
UCL
CL
LCL Rejection
Sample number
Figure 3.5: Basic structure of control chart
All the point of data value fall within this control limit then the process will be considered
as an acceptable. The point beyond this limit is unacceptable and this region is rejection
region. Although all the points fall within the limits then the process may be out of
control. Thats depends on two important criteria. The criteria which indicate that the
process is out of control are: 1) the pattern of plots of points must be random. If the last
10 point falls all below or above the central line that mean there is assignable cause
behind this. This cannot be occurred by a natural cause. 2) There must not be any
increasing or decreasing pattern, or any specific trend. If occur, there is also any
assignable cause behind.
3.1.3 Pareto Analysis
This is one of the most important tools of TQM basic seven tools. In the early nineteenth
century, the famous Italian Economist Vifredo Pareto observed and stated that about 80%
of the countrys wealth is occupied by about 20% of the population. This famous
observation was later named as 80-20 rule. Although the observation was concentrated
on only wealth distribution but later the researcher has found that this rule can be equally
applicable to the other fields of knowledge. In quality control, it has been seen that 80%
Q
u
a
l
i
t
y

c
h
a
r
a
c
t
e
r
i
s
t
i
c

v
a
l
e
A
c
c
e
p
t
a
n
c
e
13
defects or out of control situation is due to the 20% causes. If these 20% cause can be
possible to remove then the out of control situation will be reduced by 80%. This 20%
cause is known as vital few, whereas the rest many are known as Trivial many.
Pareto chart looks like a cumulative bar graph. The length of the bar represents the
frequency or cost and arranged longest bar to the left and shortest to the right. It is also
known as Pareto chart or Pareto diagram.
3.1.4 Cause-Effect Diagram
One of the famous quality gurus named Kaoru Ishikawa had introduced the cause-effect
diagram. Every adverse effect of product related to the quality characteristics which must
be result in the some specific reason or cause. Cause-effect analysis is a tool for analyzing
and illustrating a process by showing the main causes and sub-causes leading to an effect.
Drawing a Cause-effectdiagram for a quality effect with all possible causes looks like a
skeleton of a fish. For this reason its called fishbone diagram. This diagram generally
is drawn by brainstorming step by step. At first need to draw a horizontal line align with
the quality effect then possible causes are drawn to the both side of line with an arrow
line. All the sub-causes are drawn to every possible cause.
Figure 3.6: Basic structure of Cause-effect diagram
There are two types of cause-effect diagram. One is cause enumeration and another is
process analysis. Cause enumeration is the most commonly used CE diagrams in
industries. This identifies one-by-one all possible causes from brainstorming sessions and
then classifies into groups. Figure 3.6 shows the general Example of cause effect (CE)
diagram.
3.1.5 MCDM Approach
Multi criterion decision making approach is one of the popular approaches for decision
making. Decision is made depends on various criteria so the named as. There are many
methods in the category of MCDM approach. These are Analytic Hierarchical Process
(AHP), Outranking Method, Multi-Attribute Utility Approach (MAUT), Linear Weighted
Point, Judgmental Modeling, Interactive Selection Model (ISM), Categorical Method and
Fuzzy Sets. AHP indeed makes use of hierarchical structure to cope with Multi-criterion
Decision Making(MCDM)process with data from the real world.
Effect
Material
Type/grade
Method
Machine
Man
SOP
Skill
Accurac
y
14
3.1.6 Quality Function Deployment (QFD)
Quality Function Deployment (QFD) is a structured approach to defining customer needs
or requirements and translating them into specific plans of product, parts/components,
process and production plan, such that those needs are met. Voice of customer (VoC) is
the term to describe these stated or unstated customer needs or requirements. This
understanding of the customer needs is then summarized in a product planning matrix or
House of Quality (HOQ). These matrices are used to translate higher level whats or
needs into lower level hows product requirements or technical characteristics to
satisfy these needs.
3.2 Lean Manufacturing
Now-a days Lean manufacturing has become a popular waste minimization tool used in
industries. Lean manufacturing is a manufacturing strategy that seeks to produce a high
level of throughput with a minimum of inventory. Lean Production originated from the
Japanese Toyota Motor Corporation, it developed from the management of the Ford
Motor Company that Toyota Motor Corporation is studying. Its core idea is to remove all
links of enterprises of non-value-added activities, with less manpower, less equipment, in
shorter time and smaller site to create as much as possible the value to meet customer
requirements of the product and service. It emphasis on the reduction of the wastage, so
that the overall cost of production process are reduced. Lean Production follows the mass
production (mass production, MP) mode, this production mode have a greater impact to
human society, is a symbol of the new age of industrialization.
The main aspects of the Lean manufacturing are:
1. High level of throughput with minimum amount of decentralized stockpiles.
2. SMED technique to minimize setup time, which allows small lot production; even
one-of-a-kind production is possible when setup time become zero.
3. Poka-yoke technique to prevent mistake.
4. 5S Program to eliminate unnecessary materials, in order to avoid mixing up of
good with bad, and make place clean and safe.
5. Total Productive Manufacturing (TPM) and its part Total Productive Maintenance
(TPM), etc.
3.2.1 The 5S Philosophy
The 5S is one of the important tools in Lean manufacturing. Based on Japanese words
that begins with S, the 5S Philosophy focuses on effective work place organization and
standardized work procedures. 5S simplifies work environment, reduce waste and non-
value activity while improving quality efficiency and safety. 5S is the Japanese concept
for House Keeping.
The 5 Ss are:
1. Sort (Seiri) - The first S focuses on eliminating unnecessary items, removing
broken tools, getting rid of dust and oil, etc. from the work place. To identify these
unneeded items an effective visual method is used which is known as red tagging.
15
2. Set In order/ Systematize (Seiton) This is the second of the 5Ss and focuses on
efficient and effective storage methods. This phase of 5S is all about keeping
things in their rightful place.
3. Shine (Seiso) This S is concern about the thoroughly cleaning the workplace.
Daily follow-up cleaning is necessary in order to sustain this improvement.
Workers take pride in a clean and clutter free work area and shine step will help
create ownership in the equipment and facility.
4. Standardize/Systematize (Seiketon) When above 3S have been implemented
then next step is to concentrate on standardizing best practice in the work area. It
is necessary to allow employees to participate in the development of such
standards.
5. Sustain (Shitsuke) This is most difficult S to implement and achieve. Human
nature is to resist change. Sustain focuses on defining a new status quo and
standard of work place organization. Precisely speaking, the system moves into
the area of Kaizen or ongoing improvement.
3.3 The Seven Wastes in the Industrial Production
The concept of 7 wastes was popularized by Womack and Jones in their book named
The Machine that changed the world. After that TaiichiOhno in his book named
Toyota Production system explained the main foundation of Lean Manufacturing.
TaiichiOhno devised 7 categories which cover virtually all of the means by which
manufacturing organizations waste or lose money; these have become known as the7
wastes.
The 7 wasted described by Ohno are:
1. Overproduction Overproduction means producing more than customer
orders, producing unordered material/goods. It is often cause by quality
problems.
2. Waiting Products waiting around in factories either as finished goods or
work in process (WIP) another major cause of waste.
3. Transportation This type of waste concern with transportation handling
more than once, delays in moving materials, unnecessary moving or handling.
4. Inventory This type of waste is associated with the unnecessary raw
materials, work in process (WIP), & finished stocks.
5. Motion Those parts of the motion is considered as a waste which is concern
with the movement of people and equipment that does not add any value to the
final product.
6. Over-Processing Rework is a typical example of over processing as
discussed earlier reducing the root cause of the quality problem is solution
eliminating rework. Unnecessary processing or procedure that adds no value is
associated with this waste.
7. Defective Units The finished product which does not fulfill the customer
requirement or desired characteristics is known as defective products.
16
CHAPTER 4
Case Study & Data Collection
For the thesis purpose we had gone to a leading food-product manufacturing company in
Bangladesh named Pran Agro Ltd. It is situated at Natore district. There are many types
of product and individual department for the individual product. Some of the major
departments are Jam-Jelly, Spice, Choco-bean, Ice-pop, Plastic, Mango juice etc. Among
them we had chosen ice-pop department and their products for collecting data. In this
department all product are being produced in the same procedure with different shape,
color, and flavor. There were three flavored and that are Mango, orange and litchi. The
production layout of the department is shown in figure 4.1.
Figure 4.1 the production layout of ice-pop department
We have collected data related to the quality characteristics of the product such as what
types of defects normally occur, how frequently occur, what is the reason behind these,
what procedure they follow etc. There are found five types of defect named Leakage,
Black particle, without coding, Cap loose, and Short/Excess. By asking some
questionnaire about what are the causes for these defects to the worker and quality
assurance department we have found the following causes for respective defects.
Sugar syrup
Tank
Press filter
Reserve
tank
Cooking tank
Sterilizer
Cooler Dryer
Shorting
conveyor
Fine filter
fifilter
Fine filter
Keep 18-20
min at temp.
80-85 degree
to make
germ free
Check randomly
Add
chemical
17
1. Leakage
If the thickness of the tube isnot uniform
Unbalanced heat
External hit
Operators skill etc.
2. Black particle
Dirty Environment/ work area
Dirty tube
Unclear raw materials
Unclear machine.
3. Without coding
Operators insincerity
Unskillful worker
Machine etc.
4. Cap loose
Time maintaining of worker
5. Excess/Short
Operators insincerity.
From the quality assurance department it has been found the data of month of 26 working
days related to the question, what type of defects occur how frequently such as sample
size, number of non conformities, fraction non-conformities.
All data of 26 working days collected from quality assurance department are given in the
Appendix A (at the last of the paper).
18
CHAPTER 5
Calculations
5.1 Define: There are five types of defects in the product of ice-pop department of Pran
Agro Ltd. In the define phase QFD have been used to identify the relationship between
the defects and the factors that affect these defects.
5.1.1 Quality Function Deployment (QFD): Here QFD shows the relationship between
the defects and possible causes shown in figure 5.1. Here 1, 3, 9 denote
1=weak relation
3=Moderate relation
9=Strong relation
Figure 5.1 QFD model showing the relationship between defects and possible causes
In the diagram it has been seen that lack of worker attention have the highest score, then
machine setup, dust etc respectively. Here importance means is the numbering the defects
among 10.
U
n
s
k
i
l
l
e
d

o
p
e
r
a
t
o
r
R
a
w

m
a
t
e
r
i
a
l
H
e
a
t

b
a
l
a
n
c
e
M
/
C

s
e
t
u
p
D
u
s
t
N
o
i
s
e
L
a
c
k

o
f

a
t
t
e
n
t
i
o
n
M
e
a
s
u
r
i
n
g

a
c
c
u
r
a
c
y
I
m
p
o
r
t
a
n
c
e
(
1
0
)
W
e


t
o
d
a
y
(
1
0
)
T
a
r
g
e
t

i
n

f
u
t
u
r
e
(
1
0
)
I
m
p
r
o
v
e
m
e
n
t

r
a
t
i
o
S
a
l
e
s

p
o
i
n
t
S
c
o
r
e
s
P
e
r
c
e
n
t


s
c
o
r
e

(
1
0
0
)
Leakage 3 1 9 3 1 1 10 8 9 1.1 1.2 13 28
Black particle 3 3 9 3 8 6 7 1.2 1.1 11 24
Without coding 3 9 1 3 4 2 3 1.5 1.3 8 17
Cap loose 1 3 3 9 3 3 4 1.3 1.2 5 11
Short/Excess 1 1 9 3 6 4 5 1.2 1.2 9 20
Score(Sum
1664)
166 100 252 309 244 70 364 159
Percent
score(100)
10 6 15 19 15 4 22 10
19
5.2 Measure: Here to measure how many data are out of control limits P control chart
have been used.
5.2.1 Control Chart: Since the quality characteristics data is attributes type so P chart is
preferable.
Table 5.1: Data tableof non-conformities of 26 days with sample size n=810.
Sample
no
(day)
No of
abnormalities
Fraction non
conforming
Sample no
(day)
No of
abnormalities
Fraction non
conforming
1 5 0.0061 14 9 0.0111
2 4 0.0049 15 6 0.0074
3 6 0.0074 16 5 0.0061
4 9 0.0111 17 7 0.0086
5 3 0.0037 18 12 0.0148
6 6 0.0074 19 5 0.0061
7 4 0.0049 20 3 0.0037
8 11 0.0135 21 4 0.0049
9 6 0.0074 22 6 0.0074
10 3 0.0037 23 3 0.0037
11 5 0.0061 24 5 0.0061
12 6 0.0074 25 7 0.0086
13 4 0.0049 26 4 0.0049
f
Figure 5.2: P chart
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
1 2 3 4 5 6
F
N
C
19
5.2 Measure: Here to measure how many data are out of control limits P control chart
have been used.
5.2.1 Control Chart: Since the quality characteristics data is attributes type so P chart is
preferable.
Table 5.1: Data tableof non-conformities of 26 days with sample size n=810.
Sample
no
(day)
No of
abnormalities
Fraction non
conforming
Sample no
(day)
No of
abnormalities
Fraction non
conforming
1 5 0.0061 14 9 0.0111
2 4 0.0049 15 6 0.0074
3 6 0.0074 16 5 0.0061
4 9 0.0111 17 7 0.0086
5 3 0.0037 18 12 0.0148
6 6 0.0074 19 5 0.0061
7 4 0.0049 20 3 0.0037
8 11 0.0135 21 4 0.0049
9 6 0.0074 22 6 0.0074
10 3 0.0037 23 3 0.0037
11 5 0.0061 24 5 0.0061
12 6 0.0074 25 7 0.0086
13 4 0.0049 26 4 0.0049
f
Figure 5.2: P chart
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Day
19
5.2 Measure: Here to measure how many data are out of control limits P control chart
have been used.
5.2.1 Control Chart: Since the quality characteristics data is attributes type so P chart is
preferable.
Table 5.1: Data tableof non-conformities of 26 days with sample size n=810.
Sample
no
(day)
No of
abnormalities
Fraction non
conforming
Sample no
(day)
No of
abnormalities
Fraction non
conforming
1 5 0.0061 14 9 0.0111
2 4 0.0049 15 6 0.0074
3 6 0.0074 16 5 0.0061
4 9 0.0111 17 7 0.0086
5 3 0.0037 18 12 0.0148
6 6 0.0074 19 5 0.0061
7 4 0.0049 20 3 0.0037
8 11 0.0135 21 4 0.0049
9 6 0.0074 22 6 0.0074
10 3 0.0037 23 3 0.0037
11 5 0.0061 24 5 0.0061
12 6 0.0074 25 7 0.0086
13 4 0.0049 26 4 0.0049
f
Figure 5.2: P chart
21 22 23 24 25 26
UCL1
LCL1,2
UCL2
CL1
CL2
20
5.2.2 Control limits calculation:
Lets
is the fraction non-conforming. Then (1- ) is fraction conforming. Mean fraction non
conforming for the data of table 5.1 can be found from the following equation.
=0.1819/26=0.00699
Since population fraction non-conforming is not known, the value can be used to
calculate the upper and lower control limits.
UCL
1
= + 3
( )
=0.0157
CL
1
= =0.00699
LCL
1
= -3
( )
=-0.00179=0
The LCL is a negative value, which is infeasible, because fraction non-conforming cannot
be negative, thereby justifying to be taken as zero.
The control limits are drawn in figure 5.2, and subsequently 26 fraction non-conforming
values are plotted.
It has been seen in the figure that the sample no. 8 and 18 are out or so closed to the upper
control limits. So there might have some specific reason behind this and investigation
needs to identify. We can neglect the two data. After neglecting the two data of out of
control, calculate becomes
=
0.1806
24
= 0.0064
UCL
2
= + 3
( )
=0.0148
CL
2
= = 0.0064
LCL
2
= -3
( )
=-0.0084=0
Then the control limits have been plotted in the same figure. Then all data are randomly
distributed.
21
5.3 Analysis: Three tools named cause-effect diagram, Pareto chart, MCDM has used to
analyze the defects. Individual calculation shown in below;
5.3.1 Causes Effect Diagram:The possible causes for individual defects are shown in
figure 5.3(a, b, c, d, and e) respectively. The brainstorming data have been used to draw
these diagrams.
ManMachineMaterial
Supplier
Grad of
Raw material
Method Management Environment
Figure 5.3(a) Cause-effect diagram for leakage
Man Machine Material
Supplier
Method Management Environment
Figure 5.3(b) Cause-effect diagram for black particle
Leakage
Non
uniformity
Lack of attention
Temperature
Setup
Rules
Support
Noise
Safety
Dust
Temp. &
Humidity
Lack of
skill
Lack of
Training
System
Black
particle
Unclean
Lack of attention
Tooling
Setup
Rules
Support
Noise
Safety
Dust
Temp. &
Humidity
Lack
of skill
Lack of
Training
System
Composition
Shortage
Heat
Balance
Update
Manual
Manual
Update
22
Man Machine
MethodManagement Environment
Figure 5.3 (c) Cause-effect diagrams for without coding
Man Measurement
Method Management Environment
Figure 5.3(d) Cause-effect diagram for shortage/excess
Without
Coding
Lack of attention
Tooling
Setup
Rules
Support Noise
Safety
Temp. &
Humidity
Lack
of skill
Lack of
Training
System
Shortage
Accuracy
Shortage
/Excess
Lack of attention
Tooling
Time
Rules
Support
Noise
Safety
Temp. &
Humidity
Lack
of skill
Lack of
Training
System
Shortage
Accuracy
Update
Appropriatenes
s
Update
Appropriatenes
s
Manual
23
Man Machine
Method Supplier Environment
Figure 5.3(e) Cause-effect diagram for loose caps
In the above all cause-effect diagram possible causes have been divided into some major
factors by brainstorming data. Then also some specific factors related to those major
factors have beenidentified.
Loose
Caps
Lack of attention
Tooling
Setup
Responsibility
Bad
Quality
Noise
Safety
Temp. &
Humidity
Lack
of skill
Lack of
Training
Shortage
Accurac
y
Heat
Appropriateness
Manual
24
5.3.2Pareto Analysis
In our survey time, we have found among64 defective itemsthe no. of individual defects
is shown in table 5.2. The individual defects, its number and percentage of frequency are
given bellow:
Table 5.2: Different types of defects percentage
Defect Num. of abnormalities Frequency(percentage)
Leakage 31 48
Black particle 22 35
Without coding 05 8
Cap loose 04 6
Short/Excess 02 3
Figure 5.4 Pareto Chart
In the figure 5.4 shows that 48% defects is leakage, 35% black particle and remain
percents other. So here vital few is these two defects. If it possible to remove then about
80%defects will be minimized.
48
0
10
20
30
40
50
60
70
80
90
100
Leakage Black particle
Defect
F
r
e
q
u
e
n
c
y
24
5.3.2Pareto Analysis
In our survey time, we have found among64 defective itemsthe no. of individual defects
is shown in table 5.2. The individual defects, its number and percentage of frequency are
given bellow:
Table 5.2: Different types of defects percentage
Defect Num. of abnormalities Frequency(percentage)
Leakage 31 48
Black particle 22 35
Without coding 05 8
Cap loose 04 6
Short/Excess 02 3
Figure 5.4 Pareto Chart
In the figure 5.4 shows that 48% defects is leakage, 35% black particle and remain
percents other. So here vital few is these two defects. If it possible to remove then about
80%defects will be minimized.
35
8
6
3
Black particle Without coding Cap loose Short/Excess
Defect
F
r
e
q
u
e
n
c
y
24
5.3.2Pareto Analysis
In our survey time, we have found among64 defective itemsthe no. of individual defects
is shown in table 5.2. The individual defects, its number and percentage of frequency are
given bellow:
Table 5.2: Different types of defects percentage
Defect Num. of abnormalities Frequency(percentage)
Leakage 31 48
Black particle 22 35
Without coding 05 8
Cap loose 04 6
Short/Excess 02 3
Figure 5.4 Pareto Chart
In the figure 5.4 shows that 48% defects is leakage, 35% black particle and remain
percents other. So here vital few is these two defects. If it possible to remove then about
80%defects will be minimized.
3
Short/Excess
Series1
Defect
F
r
e
q
u
e
n
c
y
25
5.3.3MCDM approach: AHP methods have been used to rank these defects depending
on some factor denoted by C
1
, C
2
, C
3
, and C4 shown in figure 5.5. Here Goals denote
which one is more critical.
Figure 5.5 Proposed AHP model
Table 5.3: Level of preference weights
Level of preference
/importance
weights
Definition Explanation
1 Equally preferred Two activity contribute equally to the
objective
2 Moderately Experience and judgment slightly favor
one activity over another
3 Strong Importance Experience and judgment strongly or
essentially favor one activity over
another
4 Extreme Importance The evidence favoring one activity over
another is of the highest degree possible
of affirmation
Reciprocals Reciprocals for inverse comparison
Goals
Method,
C
1
Environment
/ Workplace,
C
2
Man,
C
3
Management,
C
4
Leakage,
A
1
Black
Particle,
A
2
Without
Coding,
A
3
Cap
Loose,
A
4
Excess
/Short,
A
5
26
Table5.4: Evaluation at level 1
Attribute C
1
C
2
C
3
C
4
Geometric
mean
Normalized
weight
C
1
1 3 1/2 4 1.57 0.3298
C
2
1/3 1 1/3 2 0.686 0.144
C
3
2 3 1 3 2.06 0.432
C
4
1/4 1/2 1/3 1 0.452 0.0949
Sum 3.583 7.5 2.167 10 4.76
Note: Here in the Table 5.4
Geometric Mean is being calculated by this Formula: [a
1
*a
2
*a
3
*.a
n
]
1/n
where a
1
,
a
2
,a
n
are the elements of any particular row and n is the number of elements in that
row.Then the Normalized Weights are being calculated as dividing each element of
Geometric mean by the summation of all the geometric means.
Example: For first row,
Geometric Mean = [1*3*1/2*4]
1/4
= 1.57
Then the Geometric Mean = 4.76
Now the Normalized Weight for first row = 1.57/4.76 = 0.3298
Sample Calculation
Eigenvector

max
=(Normalized weight of each row*sum of respective column)
= (3.583*0.3298+ 7.5*0.144+2.167*0.432+10*0.0949) = 4.1469
As follows: Consistency index CI = (
max
n) / (n1) = 0.0489; Where n is the
number of criteria are being considered. Here n= 4
Now the consistency ratio C.R = C.I/R.I = 0.0489/0.89 = 5.49%<10% so it can be said
that the data those are taken are consistent because a data to be perfectly consistent its
consistency ratio should be lies within 10%.
Where C.I means Consistency Index and R.I means Random Index. Here Random Index
(R.I) is being taken from Saatys proposed table for the value of n = 4.
Table5.5:Average Random Index (RI)based on matrix size(adapted by Saaty)
N 1 2 3 4 5 6 7 8 9 10
RCI 0 0 0.52 0.89 1.11 1.25 1.35 1.40 1.45 1.49
Table 5.6: For the factor method
Attribute A
1
A
2
A
3
A
4
A
5
Geometric
mean
Normalized
weight
A
1
1 1/4 1/3 2 1/2 0.5564 0.0947
A
2
4 1 2 3 2 2.1411 0.365
A
3
3 2 1 2 2 1.609 0.274
A
4
1/2 1/3 1/2 1 2 0.774 0.1318
A
5
2 1/2 1/2 1/2 1 0.792 0.135
Sum 11.33 3.533 3.167 7.7 6.867 5.873
27
Eigen Vector
max
=
(11.33*0.0947+3.533*0.365+3.167*0.274+7.7*0.1318+6.867*0.135) = 5.1721. For n=5;
R.I= 1.11 (From Table5.5)
Consistency Index (C.I) = (
max
-n)/ (n-1) = 0.043
Consistency Ratio (C.R) = C.I/R.I = 3.87% < 10%, so acceptable
Table 5.7: For the factor environment
Attribute A
1
A
2
A
3
A
4
A
5
Geometric
mean
Normalized
weight
A
1
1 1/3 1/3 3 2 0.8705 0.167
A
2
3 1 2 2 1/2 1.35 0.26
A
3
3 2 1 1/4 1/2 0.9028 0.174
A
4
1/3 1/2 4 1 2 1.173 0.226
A
5
1/2 2 2 1/2 1 0.903 0.174
Sum 8 4.7 9.033 6.2 6 5.2
Eigen Vector
max
= (8*0.167+4.7*0.26+9.033*0.174+6.2*0.226+6*0.174) = 5.575. For
n=5; R.I= 1.11 (From Table5.5)
Consistency Index (C.I) = (
max
-n)/ (n-1) = 0.143
Consistency Ratio (C.R) = C.I/R.I = 1.29% < 10%, so acceptable
Table 5.8: For the factor man
Attribute A
1
A
2
A
3
A
4
A
5
Geometric
mean
Normalized
weight
A
1
1 1/2 3 1/2 2 1.084 0.215
A
2
2 1 2 1/3 1/2 0.92 0.183
A
3
1/3 1/2 1 3 4 1.148 0.228
A
4
2 3 1/3 1 1/2 1.00 0.199
A
5
1/2 2 1/4 2 1 0.87 0.173
Sum 5.83 7 6.58 6.83 8 5.022
Eigen Vector
max
= (5.83*0.215+7*0.183+6.58*0.228+6.83*0.199+8*0.173) = 5.327.
For n=5; R.I= 1.11 (From Table5.5)
Consistency Index (C.I) = (
max
-n)/ (n-1) = 0.082
Consistency Ratio (C.R) = C.I/R.I = 7.36% < 10%, so acceptable
Table 5.9: For the factor management
Attribute A
1
A
2
A
3
A
4
A
5
Geometric
mean
Normalized
weight
A
1
1 1/3 1/3 3 2 0.92 0.18
A
2
3 1 1/2 2 1/2 1.084 0.22
A
3
3 2 1 1/4 1/2 0.944 0.18
A
4
1/3 1/2 4 1 2 1.059 0.21
A
5
1/2 2 2 1/2 1 1 0.199
Sum 7.83 5.83 7.83 6.75 6 5.01
Eigen Vector
max
= (7.83*0.18+5.83*0.22+7.83*0.18+6.75*0.21+6*0.199) = 5.432. For
n=5; R.I= 1.11 (From Table5.5)
28
Consistency Index (C.I) = (
max
-n)/ (n-1) = 0.108
Consistency Ratio (C.R) = C.I/R.I = 9.7% < 10%, so acceptable
Table5.10: Final Evaluation
Alternative Attribute & their weight Composite
weight
Rank
Method
0.3298
Environment
0.144
Man
0.432
Management
0.0949
Leakage 0.0947 0.167 0.215 0.18 0.165 4
Black
particle
0.365 0.26 0.183 0.22 0.257 1
Without
coding
0.275 0.174 0.228 0.18 0.23 2
Cap loose 0.1318 0.226 0.199 0.21 0.182 3
Excess/Short 0.135 0.174 0.173 0.199 0.163 5
Note:
Here Composite weight for the first row:
(0.3298*0.0947+0.144*0.167+0.432*0.215+0.0949*0.18) = 0.165
From the above result it can be said black particle has the highest Composite weight
0.257.
There is no calculation in the Improve and Control phase. So these have been discussed
in the next chapter.
29
CHAPTER 6
Result Analysis
6.1 Define:In this phase problem have been identified. QFD diagram shown in figure 5.1
represent the relation between the defects and possible causes. In the diagram it indicates
that the lack of attention of the worker has the highest relation with the defects.
6.2 Measure:Pcontrol chart have been used to measure the condition of the process
shown in figure 5.2. It shows that data on the 8
th
and 18
th
out or near the upper control
limits. There must have some specific assignable reason behind this data. Investigation
must need to do for what such type of result is coming.
6.3 Analysis: In this phase, fish-bone diagram, Pareto chart, MCDM approach has been
used to analyze. Figure 5.3 shows the possible causes for the individual defect. All the
possible causes are from the brainstorming session. It does not show the actual cause for
defects. Pareto chart have been used to identify which defects are occurring more
frequently that means vital few. Figure 5.4 of Pareto chart shows the leakage and black
particle is the vital defects. If these two defects can be possible to remove then defects
will be reduced up to 80%. In the table 5.10 of the MCDM it has been seen that black
particle is rank one defects. The causes of black particle need to be eliminated first.
6.4 Improve:In the analysis phase noticed that black particle and leakage is the main
problem. 5S philosophy can be implemented for reduce black particle since work area/
Environment is the main cause for it. Method needs to be update and operators also need
to train up to reduce leakage problem. The tubes of ice-pop department from the supplier
need to inspect properly because thickness of the tube is one of main factor for leakage.
In necessary supplier can be changed.
6.5 Control:This solution and continuous improvement process must need to maintain
over time. For this purpose continuous training schedule for the worker need to setup and
update new standard of documents (i.e. procedure, work inspection) must be established.
30
CHAPTER 7
Recommendations and Future Works
The key objective of this study was to identify the quality problem of a product of a
manufacturing industry such as defects and its possible causes. And also the application
of Lean and Six Sigma tools for the purpose of decreasing the defects. In this study,
various Six Sigma and Lean tools such as Pareto analysis, cause-effect diagram, control
chart and also AHP technique of MCDM approach, QFDhave been used. Other
techniques of MCDM approach such as Grey relational analysis (GRA), fuzzy sets may
be applied here. Data have been taken over one month only. If more data was taken it
would give more precise results. Here only defective items and their causes have been
described and have tried to overcome these. Other types waste such as motion; inventory,
transportation etc. also can be solved by this technique.Only p chart have been used to
measure the problem, other type likes u, c, npetc, also can be applied and use of more
than one would more precise results. Value process map can be used for the purpose of
identification the activity which does not add value. Non value added activity will need to
be identified to apply 5S philosophy or elimination of other type of defects. But here non
value added activity have not identified although 5S tools have been suggested to
evaluate the work place area. This is one of the lacks of this study. If there is relation
between the defects to each other or dependency then regression analysis may be used to
understand the relationship.
In the future work, implementation of Lean Six Sigma can be carried out other type of
waste like motion, inventory, transportation etc. The application of Lean Six sigma in the
service sectors is inconsiderable yet now. This may be another further research objective.
31
CONCLUSION
Six sigma is a powerful method that gathered many important aspects of various tools to
improve the process. Lean and Six Sigma both have been implemented as integrated form
in this study to obtain better results and support to each others. Lean Six Sigma can be
applied easily in any kind of business areas likes service, production, marketing, sales,
procurements etc. The major advantages are reduced cost, reduced time, maximization of
profits, quality of the products, increased customer satisfaction etc. The factors which
influence the successful implementation of Lean Six Sigma are management
involvements, organizational commitment, control skill, continuous training and cultural
change. In this thesis work, Lean Six Sigma have been implemented and recommended
for decreasing the defects of products, ultimately for reducing the process variation.
Although all preventive and corrective action has defined but for this moment it cannot be
said that the number of defective items is decreasing because till now no further analysis
have been performed. But it can be assured that the successful implementation of this
management technique must be beneficial to the organization.
32
REFERENCES
1. Amer, Y, Loung, L, Lee, SH &Asharf, MA 2008, Optimizing order fulfillment using
design for six sigma and fuzzy logic, International Journal of Management Science and
Engineering Management, Vol.3, No.2, PP. 83-99.
2. Atmaca, E &Girenes, SS 2011, Lean six sigma methodology and Applications, The
Red Crescent Society, Turkey.
3. Barac,N, Milovanovic,G&Andjelkovic,A 2010, Lean production and six sigma quality
in lean supply chain management, Economics and Organization, Vol.7, PP.319-334.
4. Brun, A 2011, Critical success factors of six sigma implementations in Italian
companies, Int.J. Production Economics, Vol.131, PP.158-164.
5. Chakravorty, SS 2009, Six sigma programs: An implementation model, Int.J.
Production Economics, Vol.119, PP.1-16.
6. Chang, KK & Wang, FK 2008, Applying Six sigma methodology to collaborative
forecasting,Int J AdvManufTechnol, Vol.39, PP.1033-1044.
7. Cudney,E.A, Mehta M & Monroe R 2006, Combining lean and sox sigma for
opticalresults, Society of manufacturing engineers, California.
8. Ditahardiyani, P, Ratnayani, &Angwar, M 2008, The quality improvement of primer
packaging process using six sigma methodology, JurnalTeknikIndustri, Vol. 10, No.2,
PP.177-184.
9. Does, RJ &Koning, HD 2006, Lean Six Sigma in a hospital, Int. J. Sigma and
Competitive Advantage, Vol.2, No. 2.
10. Fairbanks, CB 2007, Using Six Sigma and Lean methodologies to improve OR
throughput, Aorn Journal, Vol.86, No.1.
11. Hasin, AA 2007, Quality Control and Management, 1
st
edition, Bangladesh Business
Solutions, Dhaka, Bangladesh.
12. Hekmatpanah, M, Sadroddin, M, Shabaz, S, Mokhtari, F &Fadavinia, F 2008, Six
Sigma process and itsimpact on the organizational productivity, World Academy of
Science, Engineering and Technology, Vol. 43.
13. Hwang, YD 2006, The practice of integrating manufacturing execution systems and
six sigma methodology, Int J AdvManufTechnol,Vol.31, PP.145-154.
14. Johannsen,F, Leist, S &Zellner, G 2011, Six sigma as a business process
management method in services: analysis of the key application problems, InfSyst E-Bus
Manage, Vol.9, PP.307-332.
33
15.Khalil,MS,Khan,MA&Mahmood T 2006, Lean Six sigma A tool to improve
productivity, quality and efficiency in manufacturing and industrial sector, Pakistan.
16. Kwak, YH &Anbari, FT 2006, Benefits, obstacles, and future of six sigma
approach, Technovation, Vol. 26, PP. 708-715.
17. Mast,Jd&Lokkerbol, J 2012, An analysis of the six sigma DMAIC method from the
perspective of problem solving, Int.J. Production Economics, Netherlands.
18. ONeill, DM & Duvall, C 2003, A Six Sigma quality approach to workplace
evaluation, Journal of Facilities Management, Vol.3, No. 3, PP. 240-253.
19. Padhy, RK &Sahu, S 2011, A real option based Six Sigma project evaluation and
selection model, International Journal of Project Management, Vol. 29, PP. 1091-1102.
20. Yang, T & Hsieh, CH 2009, Six-Sigma project selection using national quality
award criteria and Delphi fuzzy multiple criteria decision-making method, Expert
Systems with Applications, Vol. 36, PP. 7594-7603.
34
Appendix A
Data table of the defects of ice-pop in Pran Agro Ltd.
Date Sample
size
Leakage Black
particle
Without
coding
Loose
caps
Excess/
shortage
Total
01.04.2012 810 2 1 0 1 1 5
02.04.2012 810 2 1 0 1 0 4
03.04.2012 810 3 2 1 0 0 6
04.04.2012 810 5 2 1 1 0 9
05.04.2012 810 1 0 0 1 1 3
07.04.2012 810 2 2 1 1 0 6
08.04.2012 810 1 1 2 0 0 4
09.04.2012 810 4 4 2 0 1 11
10.04.2012 810 2 2 1 1 0 6
11.04.2012 810 1 1 0 0 1 3
12.04.2012 810 2 2 1 0 0 5
14.04.2012 810 3 1 1 1 0 6
15.04.2012 810 1 2 0 0 1 4
16.04.2012 810 4 2 1 2 0 9
17.04.2012 810 2 2 2 0 0 6
18.04.2012 810 2 2 1 0 0 5
19.04.2012 810 3 1 1 1 1 7
21.04.2012 810 3 6 2 0 1 12
22.04.2012 810 3 1 1 0 0 5
23.04.2012 810 0 1 2 0 0 3
24.04.2012 810 1 1 0 1 1 4
25.04.2012 810 2 1 2 1 0 6
26.04.2012 810 2 1 0 0 0 3
28.04.2012 810 2 2 1 0 0 5
29.04.2012 810 3 1 1 1 1 7
30.04.2012 810 1 2 1 0 0 4