Quality Function Deployment (QFD): –
is focused methodology which takes into consideration the customer
requirements, needs and provides technical solutions to their needs. During
this the QFD also establishes the inter-relationship between the technical
solutions. It also does a quick comparison and bench marking amongst the
helps decision maker to understand what are the target values and what
improvements should be done to achieve the target values. It also compares the
technical solutions with each other thus giving better idea for satisfaction of
QFD methodology can be applied to tangible as well as non-tangible services and
of Quality is a key component of QFD.
Structure of House of Quality: –
house of quality involves the following parts:
Voice OF Customers
– The QFD is aimed at voice of customers. So, the house of quality starts with
the voice of customers on the extreme left. The customer needs and requirements
give ‘What’ problems must be fulfilled.
Technical Descriptors- Technical
descriptors are placed on the top. They are voice of engineers and give the
technical solutions to satisfy the customer need. These part gives ‘How’ the
customer needs can be satisfied.
Inter-relationship Matrix- It is placed on the top of the Technical descriptors
and is also known as the ‘Roof’. It gives inter-relations between the technical
solutions. It helps engineers to decide how the technical solutions are related
to each other and check if there are any conflicts between the technical
descriptors aimed to solve different needs.
Relationship Matrix – It is placed between the voice of customers and the
technical descriptors. It gives the relationship between various needs and
Competitive benchmarking – It is placed on the right. It compares various
competitive products based on what target value should be reached and how all
the products are closer to that. It also has importance to voice of customer
which decides how much importance should be given to a particular customer
Technical Competitive assessment – It is placed to
the bottom of house of quality. It indicates which technical descripts work
well to satisfy the customer needs. It is based on the absolute and relative
values obtained from the house of quality. It yields overall importance rating
to the technical descriptors
The carious parts of the
house of quality can be summarized in the figure below:
Figure 1: House of Quality
QFD for Fuel Injector:
Fuel Injector: – It
is a part of fuel injection system. It is a crucial component in engine
performance. It consists of a nozzle body, spring, cap nut and nozzle.
The current study is mainly aimed at the nozzle.
Figure 2: Schematic view of fuel injector
The main customer of the
Fuel injector is engine manufacturer. The requirements/ needs described by him
was cascaded in the Voice of customer on the left of house of quality.
Voice of Customer:
The automaker companies
demand that the fuel injector nozzle should have following properties-
elevated temperature stability
high compressive stresses
voice of customers is incorporated in the House of Quality as shown in the
Figure 3: Voice of Customers
The engineers studied the
customer needs and to address the needs they have selected the following
a). Material selection
High carbon steel
c). Surface finishing
The voice of engineers or
technical descriptors are cascaded in to House of Quality as shown below:
Figure 4: Technical Descriptors
Technical descriptors are weighed against Voice of
customers as strong, medium, and weak. These later helps in technical
The technical descriptor which can satisfy the
particular customer need to higher extent are weighed as high. And the one
which can satisfy customer need to a lower extent are weighed as weak.
Figure 5: Relationship Matrix showing relation between
technical descriptors and Voice of Customer.
selected Technical Descriptors are weighed against each other. The
inter-relationship matrix indicates how the technical descriptors are related.
For example, Titanium (material) and Grinding (finishing process) have strong
positive relation. So that they can be used in conjunction with each other.
While PM with inserts and electrolyte treatment have negative relation. So,
they shouldn’t be used together. The technical descriptors which don’t conflict
with each other should be selected.
Figure 6: Inter-Relationship Matrix showing relation
between technical descriptors (Roof).
The fuel injectors from
two other companies A and B are selected for the comparison with our product.
Company A manufactures fuel injector nozzles with Aluminum using Powder
Metallurgy with insets and are surface finished by electrolytic treatment.
While company B uses High
carbon steel with Laser cutting and is finished by Grinding.
PM with Inserts
High Carbon steel
Table 1: Competitors and Reference product description
Customer Competitive Assessment:
Each customer need is weighed. The customer
requirement is directly related to target value. The products are compared on
how closely they can achieve target value.
The technical descriptors are rated with the help of
relationship matrix. The customer needs with their importance are multiplied to
each of the relation values (strong, medium, weak) in the relationship matrix
and Absolute weight is calculated. Relative weight is calculated using absolute
weight from customer market evaluation and relation values from the
The material, manufacturing process and finishing
process with the highest absolute weight in that category is selected as the
best technical solution.
House of Quality for Fuel Injector Nozzle:
Figure 7: Complete House of Quality (QFD) with best
technical solutions highlighted in red.
The house of quality or QFD indicates that the
Titanium is the best material, Laser cutting is the best process and Grinding
is the best surface finish method for the fuel injector nozzle. (depending upon
the absolute values)
From the QFD, our company needs to change to Titanium
Fuel Injector nozzles manufactured by Laser cutting and surface finished with
Six Sigma is statistical
based methodology originally developed by Motorola in 1980. It is an approach
towards improvement of product, productivity and quality and also reducing the
operational cost. It focuses on reduction of variation. Six Sigma terminology
denotes ‘Sigma Level’ for performance a company. The higher sigma level
indicates the defects in PPM are less.
means the company had 3.4 PPM defects only. Six Sigma methodology was
originally implemented for the manufacturing process. But today it can be
extended to logistics, purchases, human resource. With the help of Six Sigma
the new business models can be developed for the betterment of the company. Starting
form business strategy till the complete final implementation the Six Sigma
will result in increased profit for the company. It allows companies to
drastically improve their performance by designing and monitoring everyday
business activities in such way so that the wastes and resources can be
minimized without hampering customer satisfaction.
the important Six sigma’s approach is DMAIC. The same approach is used for the
case study of the grinding process done subsequently.
approach has been followed to reduce the process variation and the associated high
defect rate. DMAIC is an acronym for Define, Measure, Analyze, Improve and
description of each of the phase is as follows:
This is the first phase of DMAIC.
In this phase the problem is identified, and project scope and boundaries are
defined. The customer requirements and expectations are defined. Ultimately the
goal is set for a process. This phase involves team work and generally top
management is also involved.
In this phase the measurement factors which are to be improved are
selected. A structure to evaluate current performance and assessing, comparing
and monitoring subsequent improvements and their capability is provided.
In this phase, data collected from
the Measure phase is analyzed using analysis tools such as quality tools (Fault
Tree Diagram, Fish-Bone Diagram, Pareto Chart) and the root cause is
identified. This phase helps in understanding the cause and it compares and
prioritizes the opportunities for betterment.
This phase aims at the elimination
or reduction of the root cause of variation. The solutions are found and
implemented for the root causes identified form the Analyze phase. Generally
experimental and statistical tests for the solution are done in this phase. New
ideas are developed, tests are done and if the applied condition reduces waste
or variation it is standardized.
is the last phase of DMAIC. It ensures that the improvements are sustained. The
ongoing performance is monitored, and the process improvements are documented.
It is very important to establish standard measures to maintain performance. If
the control phase is not implemented properly, the improved process may go back
to its previous state.
The DMAIC process can be summarized
in the figure below:
Figure 8: DMAIC Approach.
The case studied is of the
reduction of the defects in a fine grinding process in the Fuel Injector
manufacturing company. The defects are reduced or eliminated with the help of
Six Sigma methodology with DMAIC approach.
The company for which the study is
done is in India. The company manufactures the whole CRDI (Common Rail Direct
Injection) systems for cars, trucks and buses. It has the manpower of about
DMAIC Approach for Case Study:
Phase 1: Define-
the team was selected for the project. Then the fine grinding process was
selected as the Critical to Quality (CTQ) characteristic for this project. The
goal was decided to reduce the rejection by 50% from the existing level, which
would ultimately result in the scrap reduction and cost saving. A SIPOC (Supplier–Input–Process–Output–Customer)
analysis was done with the help of basic flowchart for the better understanding
of the process. SIPOC analysis which is similar to the process mapping, gave a clear understanding of the process steps
needed to create the output of the process.
Phase 2: Measure-
CTQ was the rejection percentage of distance pieces after the fine grinding process.
These rejections were mainly due to the occurrence of several types of defects
like burr, shades, deep lines, patches and damage, on the component after grinding
process. The different defects are shown in the figure below:
Figure 9: Various defects in Fine grinding component.
these defects can be visually inspected. Master
samples were provided for identifying each of these defects.
collection plan was prepared with details of types of data, stratification
factors, sampling frequency, method of measurement, etc. for the data to be
collected. The data collection plan thus prepared is as below:
Table 1: Data collection plan.
the defined time domain of data collection, 368219 components were inspected,
and 61198 components were found to be rejected due to various defects. The
rejected components were having one or more defects.
collected data showed that the rejection in the process was 166200 PPM. The
corresponding sigma rating of the process can be approximated to 2.47.
goal was set to reduce the rejection by 50%, the targeted rejection at fine
grinding process was reduced to 83 100 PPM from the existing level of 166 200
Phase 3: Analyze-
collection of the data, the different defects were graphically represented in
the Pareto Chart as below:
10: Pareto Chart of defects.
brain-storming session was carried out to find the different causes of the
rejection. A cause-effect diagram was produced through this session. The
Cause-Effect diagram is shown in the Figure 11 below.
potential causes were studied to obtain the root causes of the defects. The
type of analysis was discussed to decide if the potential cause was root cause
11: Cause-Effect Diagram for Rejection
causes such as supplier to supplier variation was checked with ANOVA and it the
‘p’ value obtained (0.407>0.05) showed that it was not significant. The
various machine parameters were checked and their relationship with each other
and toward the defect was studied with DOE.
causes which include variation were calculated using the statistical approach,
while causes such as ‘Improper cleaning after dressing’ etc. were studied with
the help of GEMBA.
following table gives an idea about which potential cause was analyzed by which
method and how many of them were root causes.
2: Validation of causes
Phase 4: Improve-
was planned for optimizing the machine parameters like load applied, initial
coolant flow rate, upper wheel rpm, lower wheel rpm and cage rpm. These machine
parameters were decided based on the brain-storming session amongst the team
members and the operators.
decided to do the experiments at three levels since the relation with Material
Removal Rate (MRR) was not known. The levels selected for the machine
parameters are as shown in the table below:
3: Process parameters and their levels. (* indicates current level)
decided to estimate the effect of upper wheel rpm, load applied with lower
wheel rpm and load applied with cage rpm three interactions also because the
brain storming showed possibility of their interaction. Six parameters at three
levels and three interactions require high number of experiments which was not
possible. Hence the effects of parameters were calculated using the 27
experiments with the help of Orthogonal Array (OA).
was conducted with six parameters and three interactions, L27(313)
orthogonal array. The columns of this array are mutually orthogonal. The
response of the experiment was decided as material removal rate (MRR).
design layout prepared as per L27(313) orthogonal array is
as shown in Table 4 below:
4: Design Layout of Orthogonal Array
experiments were conducted according to the design layout and the data was
collected. This data was analyzed by Taguchi’s Signal-to-Noise (S/N) ratio
method. The S/N ratio helps in the process improvement and reduction in the
ratio is an output from an experiment, which is a measure of the variation when
the noise is present in the system. Since MRR should be constant irrespective
of the work piece, S/N ratio of ‘nominal the best’ type was selected. It was
defined for each of the experiment with the characteristics of 10?log((Y 2)/
s2), where .Y is the average and s, the standard deviation.
performed for 27 S/N ratios obtained from the experiments and it showed that effect
between load applied and upper wheel rpm was significant at 5% level of
S/N plots were obtained. The highest value of each S/N parameter showed best
level for each parameter.
12: Main effects plot for S/N ratios
The optimum level selected for each
parameter was taken as the solution for the root causes.
The table below shows the optimum
level for the parameters:
5: Optimum combination of parameters
The root causes and their solutions
are summarized in the table below:
6: Root causes and respective solitons
After obtaining the optimum
parameters it was necessary to do risk analysis. The risk analysis showed no
significant negative effect of each optimized parameter on each other.
All the solutions obtained were
implemented and the results obtained were studied. The graph of rejection of
the parts before and after shows the improvement in the process. The results
are as shown below:
13: Results: rejection percentages before and after
The controlling of the process is
the most difficult phase. But the control phase was carried out pretty good
with the help of Quality plans and control plans. The CTQs of the projects were
added to the internal audit checklist and the verifications were performed
during the audits.
The U chart was introduced for
monitoring the process limits.
The data collected after
implementation of the process showed that the reject rate was fallen to 1.19%
which previously was 16.6%. The sigma value of the process was increased to
3.76 from the previous value of 2.47. After cost analysis it was found that the
annual cost saving was around $2.4 million.
The case study shows us how the Six
Sigma methodology using DMAIC can be fruitful and yield a better cost reduction
in long run. Due to application of DMAIC the cost associated with rejection,
repair, scrap, re-inspection and tool can come down drastically with the annual
saving as high as $2.4 million.
The rejection rate was also reduced
significantly from 16.6% to 1.19%. The importance of DOE can also be
experienced with this case study.