Journal of Manufacturing Engineering, March 2018, Vol. 13, Issue.

1, pp 042-050

1 SIX SIGMA DMAIC METHODOLOGY IMPLEMENTATION IN

2 AUTOMOBILE INDUSTRY: A CASE STUDY
*Raman Sharma1, Pardeep Gupta2 and Vipin Saini3
1,3
Department of Mechanical Engineering, Seth Jai Parkash Mukand Lal Institute of Engineering and Technology,
Yamunanagar-135001, Haryana, India
2
Department of Mechanical Engineering, Sant Longowal Institute of Engineering and Technology, Longowal,
Sangrur- 148 106, Punjab, India

ABSTRACT
This paper deals with the applications of Six Sigma, DMAIC methodology to improve the
Sigma level of the project taken from an automobile industry. Modern manufacturing industries are
focusing on many innovative techniques and management practices such as Six Sigma, total
productive maintenance (TPM), total quality management (TQM), just in time (JIT), enterprise
resource planning (ERP) etc. Six Sigma offers a unique approach that is widely used in industries in
order to improve the process and reduce the number of defects. Six Sigma is a fact-based, data-driven
philosophy and methodology that improves quality by analyzing data with the help of statistics to find
the root cause of quality problems and control by preventing defects. Six Sigma provides 1.5 Sigma
drift margin from the process mean to either side, so that final products would be 99.97% defect free,
having 3.4 DPMO. One Sigma gives a precision of 68.27%, two Sigma gives 95.45% and three Sigma
of 99.73%, whereas Six Sigma gives a precision of 99.9997%.The DMAIC (define-measure-analyze-
improve-control) approach has been followed to solve an underlying problem (To reduce the in-house
rejections of Cushion P-70 Bolt RR) of reducing process variation and the associated high defects
rate. This paper explores how an automobiles industry can use a systematic and disciplined approach
to move towards Six Sigma quality levels. The DMAIC phases are utilized to decrease the defect rate
of Cushion P-70 Bolt RR (Splendor bike Shock Absorber attachment bolt) from 121550 PPM to 4263
PPM and increased in Sigma level from 2.67 to 4.11. The Process Yield increased to 99.6% from a
very low level of 87.8% and Process Capability increased to 1.93.

Keywords: Six Sigma, DMAIC, DPMO/PPM, Process Capability, Cushion P-70 Bolt RR

1. Introduction
Six Sigma is a statistical term that measures difference between 99.73% and 99.9997% quality
how far a given process deviates from perfection. “Six levels [10].
Sigma is a fact-based, data-driven philosophy and
methodology that improves quality by analyzing data Table 1. Sigma levels and DPMO [11]
with the help of statistics to find the root cause of
quality problems and control by preventing defects”. Cost of
Process
Six Sigma provides 1.5 sigma drift margin from the Capability
DPMO Performance Quality
process mean to either side, so that final product would %
be 99.97% defect free, which have only 3.4 defects per 2 308537 69.1% > 30
million opportunities as given in Table1 [12]. 3 66807 93.3% 15 – 25
4 6210 99.4% 10 – 15
5 233 99.7% 5 – 10
Six Sigma is a business strategy and a 6 3.4 99.997% <5
philosophy of working smarter not harder. One Sigma
level gives a precision of 68.27%, two Sigma gives
95.45% and three Sigma of 99.73%, whereas Six 1.1 Six Sigma in Normal Distribution
Sigma gives a precision of 99.9997%. Although To achieve 6-Sigma Quality, a process must
99.73% sounds very good quality level but it slowly produce no more than 3.4 defects per million
dawned on companies that there is a tremendous opportunities, refer Fig.1 . An opportunity is defined as

*Corresponding Author - E- mail: raman1991sharma@jmit.ac.in

www.smenec.org 42 © SME
 Journal of Manufacturing Engineering, March 2018, Vol. 13, Issue. 1, pp 042-050

a chance for nonconformance, or not meeting the effective use of six sigma methodology to reduce waste
required specifications [12]. in individual production. Six sigma DMAIC
Methodology was carried out. Company decided to
reduce detected non conformities at final inspection
from 33% to 10% and to move from 1 sigma level to 3
sigma level. Hongbo Wang , 2009 [5] summarized four
issues within the sub-category of the initial Six Sigma
concepts: basic concept, DMAIC, DFSS and
deployment. Some sectors that benefit from the
implementation of Six Sigma are listed out, and the key
factors influencing the successful Six Sigma project
implementation are identified. Tushar N. Desai et.al.
2008 [6] presented the quality and productivity
improvement in a manufacturing enterprise through a
case study. Six Sigma (DMAIC) improves the process
performance (process yield) of the critical operational
Fig.1 Sigma variations in N.D.C. process, leading to better utilization of resources,
decreases variations and maintains consistent quality of
DPMO = Number of defects * 1000000 / the process output. A.K. Sahoo et.al.2007 [7]
Number of opportunities for error per unit * Number of implemented DMAIC based Six Sigma approach to
unit optimize the operation variables of a radial forging
This means one needs to be nearly flawless in operation. M. Soković et.al, 2006 [8] presented a Six
executing key processes. The process and culture is Sigma project, undertaken within company for the
conditioned for zero defects rather than being one that production of automotive parts, which deals with
accepts that it is unavoidable, and acceptable, that identification and reduction of production cost in the
mistakes will occur. Hence Six Sigma delivers de-burring process for gravity die-castings and
substantial cost reductions, enhanced efficiencies, improvement of quality level of produced parts. M.
sustainable improvements and increased stakeholder Sokovic et.al. 2005[9] modified process design flow of
value. compressor housing with incorporate applications of
process map and the cause and effect matrix, a
2. Literature Survey comparison of the old and the modified process design
flow is made and the obtained results are discussed.
Prakash et.al.2012 [1] presented the findings
of an initial survey conducted in Indian industries. It 3. Case study
was experienced that complete implementation of Six
Sigma and its sustenance is difficult, helps A real case from a XYZ automotive industry
organizations in making right preparations for is studied. To reduce the in-house rejections of Cushion
successful implementation of Six Sigma. V.Arumugam P-70 Bolt RR (shown in Fig. 2) and improve process
et.al.2012 [2] proposed an integrated model to explain performance with the applications of Six Sigma is
process improvement implementation success through selected as a process consideration.
two learning activities undertaken by Six Sigma project
teams: Knowing-what and Knowing-how. Three
hypotheses are proposed in the model were tested using
the data collected from 52 Six Sigma project teams
from a single organization. Natha Kuptasthien et.al.
2011[3] demonstrated the implementation of Six Sigma
technique and DMAIC improvement methodology into
a mass manufacturing of printed circuit cables. The
result showed that by following the theoretical Six
Sigma technique and DMAIC steps, the defects from
major tombstone capacitor problem could be reduced
from 1,154 DPPM to 314 DPPM and increased 1st
yield output from 98.4% to 99.66%. Plecko, A. et.al,
2009 [4] presented a real case study illustrating the Fig. 2 Assembly parts of two Wheelers

www.smenec.org 43 © SME
 Journal of Manufacturing Engineering, March 2018, Vol. 13, Issue. 1, pp 042-050

have been rejected in-house inspection. The objective

Then Six Sigma DMAIC methodology is used of the project is to reduce the rejection rate of Cushion
to overcome the problems. DMAIC used the quality P-70 Bolt RR at 4000PPM level.
tools and statistics for solving the problem in different
phases like Define, Measure, Analyze, Improve and 3.1.3 Drawing of Cushion P-70 Bolt RR
Control. Fig. 4 shows the drawing of Cushion P-70
Bolt RR and all the dimensions like diameter, length,
3.1 Define Phase threads length and chamfer provided in the bolt with
During the Define phase three tasks (Fig.3) respective tolerances under which the Cushion P-70
must be undertaken as project scope, project goals and Bolt RR is acceptable.
estimating the project time and hard savings. The
purpose is to identify the problem, SIPOC analysis,
define critical customer requirements, and prepare the
team to be an effective project team.

Fig. 4 Drawing of bolt

3.1.4 SIPOC analysis (supplier, input, process,

output, customer)
Fig. 5 describes the transformation process of
Fig. 3 Current PPM to Target PPM inputs form suppliers to output for customers and gives
3.1.1Problem statement a high level understanding of the process, the process
The bolt acceptance is only 87.8% rest are steps (sub processes) and their correlation to each
rejected due to dimensional variations viz. overall other.
length of the bolt, threads cutting and chamfer
variations, collar side diameter and bolt length. The
target is set to reduce the rejections of Cushion P-70
Bolt RR at Zero % rejection level. During
manufacturing of Cushion P-70 Bolt RR, 1220
pcs./month out of avg. production of 10000 pcs./month
have been rejected in in-house inspection. This will be
a step towards achieving a target of 4000PPM
rejections.

3.1.2 Goal statement

The bolt acceptance is only 87.8% rest are
rejected due to dimensional variations viz. overall
length of the bolt, threads cutting and chamfer
variations, collar side diameter and bolt length. The
target is set to reduce the rejections of Cushion P-70
Bolt RR at Zero % rejection level. During
manufacturing of Cushion P-70 Bolt RR, 1220
pcs./month out of avg. production of 10000 pcs./month Fig. 5 SIPOC Analysis

www.smenec.org 44 © SME
 Journal of Manufacturing Engineering, March 2018, Vol. 13, Issue. 1, pp 042-050

3.1.5 Customer CTQ Requirements

The customer data (VOC) revealed that
internal customers are mainly affected by the rejections
of Cushion P-70 Bolt RR. CTQ tree shown in Fig.6 is
prepared on the basis of the VOC and project objective.

Fig. 6 C.T.Q. Tree

3.2 Measure Phase

This phase deals with the detailed process
mapping, Data collection chart, evaluation of the
existing system, assessment of the current level of
process performance, Process Flow Diagram (PFD),
Process Capability analysis and variations in the Fig. 7 Process Dimensional Variations
current process due to dimensional variations etc.
Data collected is a continuous type of data 3.2.3 Process Capability
(variable) as shown in appendix attached in the last of Process capability is commonly measured in
paper, In the case of continuous data, the data terms of the capability index (Cp), which is a ratio
collection forms are made simple and have clear space without units in Fig. 8. The purpose of this index is to
for entering the collected numerical data, the same data assess whether a process, given its usual short-term
need to be converted as discrete data by marking ok/not variation, can meet established customer requirements
ok to calculate the Sigma level of the process or specifications. Cp is a ratio of the tolerance width to
performance. the short-term spread of the process You are basically
dividing the performance standard (USL - LSL) by the
3.2.1 Sigma level calculation process width. Cp = USL – LSL / 6 x Standard
DPMO = number of defects X deviation (σ).
1000000/number of units X opportunities per unit is
given Table 2..

Table 2 Sigma Level Calculations

Sigma level calculation

PARAMETERS VALUE
No. of units 50
Opportunity per unit 1
Total no. of opportunities 50
No. of defects 6
DPMO 120000
Sigma level 2.67

3.2.2 Process Variation

With the help of collected data of the current
process we analyze dimensional variations in different
Fig. 8 Process Capability of Current (old) Process
processes and is presented in Fig.7

www.smenec.org 45 © SME
 Journal of Manufacturing Engineering, March 2018, Vol. 13, Issue. 1, pp 042-050

3.3 Analyse Phase  Severity: Quantifying the strictness of the effects

how the failure would affects the customer both
3.3.1 Cause and Effect analysis internal and external.
Cause and Effect diagram shown in Fig. 9 is  Occurrence: Likelihood of the failure occurring
mainly used tool during analyze phase since it helps based on the data and measurement.
identify the cause of a problem.  Detection: The probability of the failure being
detected before the impact of the effect is realized.

3.4 Improve Phase

3.4.1Statistical Process Control

Statistical Process Control is used to check the
process variations. With the help of MINITAB
software we calculate the individual control charts
shown in Fig.10 for the continuous data collected and
if any sample moves out of the USL or LSL then it
indicates that process is going out of control either to
stop the process or need to improve and check the fault
wherever it comes. It is a very beneficial tool it alone
can improve the process.

Fig. 9 Cause and Effect Diagram

A Cause and Effect diagram for Cushion P-70

Bolt RR presents a chain of causes and effects, sorts
out causes and organizes relationship between
variables.

3.3.2 Process Failure Mode and Effects

Analysis (FMEA)
The major causes, prioritized on the basis of
RPN, are dimensionally variations, material rusty,
chamfer uneven, thread U/S and O/S, burr, thread
damage etc.; which are responsible for rejections of
Cushion P-70 Bolt RR.

Risk Priority Number =

Fig. 10 Individual Control Charts
(Severity*Occurrence*Detection)
RPN = (S * O * D)

www.smenec.org 46 © SME
 Journal of Manufacturing Engineering, March 2018, Vol. 13, Issue. 1, pp 042-050

Table 3. Process FMEA

Sr.no. Process Potential Potential effects Potential cause S O D RPN

function failure mode (S*O*D)

Material rusty Aesthetically poor Long time storage 5 3 3 45

1 Storage Bend Inconvenience to Tool blunt and 4 3 3 36
(BOP) next operation material not put in
safe area
Dimensionally Material Without dispatch 7 4 3 84
variation rejection/supplier from supplier
2 parting rej.
Dimensionally Material Without dispatch 7 4 3 84
variation rej./supplier rej. from supplier
3 facing Dimensionally Material Without dispatch 7 4 3 84
variation rej./supplier rej. from supplier
4 Rough Dimensionally Material Without dispatch 7 4 3 84
turning variation rej./supplier rej. from supplier
Chamfer uneven Aesthetically poor Operator 8 3 3 72
5 Final negligence
turning Length not Fitment problem Operator 7 2 3 42
proper negligence
Profile NG Fitment problem Thread tool wear 4 3 3 36
out
6 Side Dimensionally PPM high ranking Without dispatch 7 4 3 84
turning variation from supplier
7 Thread Thread Inconvenience to Tool blunt and 4 3 3 36
rough next operation material not put in
safe area
8 Thread Thread U/S and Inconvenience to Operator 4 3 5 60
final O/S fitment negligence and
drill not sharp
9 chamfering Chamfer uneven Aesthetically poor Operator 8 3 3 72
negligence
Not qualify in Reject m/c setting not ok, 7 3 4 84
final R/G unskilled operator
10 Gauging Burr, thread Material reject Tool blunt and 5 3 3 45
+visual damage /rework/in customer material handling
inspection rej. ppm rank not proper
increases

www.smenec.org 47 © SME
 Journal of Manufacturing Engineering, March 2018, Vol. 13, Issue. 1, pp 042-050

3.4.3 Sigma Level Improvements

3.4.2 Process Capability Improvement
Table 4. Sigma Level Improvements
I Process Capability of overall length
dimension SIGMA LEVELS IMPROVEMENTS
Overall capability increases from 1.09 to
1.93. Before
Parameters After
Restoration
Restoration
No. of units 50 50
Opportunity per 1 1
unit
Total no. of 50 50
opportunities
No. of defects 6 0.2
DPMO 120000 4000
Sigma level 2.67 4.11
II Process Capability of bolt length
Capability increases from 0.84 to 1.50. Defects per million opportunities decreases
from 120000 PPM to 4000 PPM. And therefore
Sigma level increases from 2.67 sigma to 4.11 sigma
approx as given in Table 4..

3.4.4 Process Yield Improvements

The bolt acceptance is increased from 87.8%
to 99.6%. After the implementation of Six Sigma
project recommendations dimensional variations are
overcome by modifying the process flow instructions
and using condition monitoring of tools and
inspection gauges and chamfer oversize/undersize
III Process Capability of threads cutting in problem was also recovered. Pareto chart illustrates
the bolt reduction in rejections of Cushion P-70 Bolt RR and
Capability increases from 0.72 to 1.28 also represents the improvements in the process after
restoration of the process.
By using Minitab software capability
analysis of collected data represents that the
rejections level of Cushion Bolt P-70 reduced from
121550 PPM to 4263 parts per million as in Fig.12..

IV Process Capability of collar dimension

Capability increases from 0.81 to 1.92

Fig. 12 PPM and Yield improvements

Fig.11 (I,II,III,IV) Process Capability

Improvement

www.smenec.org 48 © SME
 Journal of Manufacturing Engineering, March 2018, Vol. 13, Issue. 1, pp 042-050

3.4.5 Product Design Improvements Overall Improvements

Rejection percentage on daily basis is

reduced from 12.2% to 0.4% as shown in the figure
4.20. It is a big achievement with the application of
Six Sigma methodology and using various Six Sigma
quality tools. The design improvements are presented
in Fig.13 and Fig. 14 and standard operating
procedure for the improved results are given table 5.

Figure 13 Design Improvements

Table 5. Standard Operating Procedure

Sr.No Operations Machines Parameters Specifications Tolerences Inspections

1 parting lathe Parting length 88.7 ±0.7 V.C

Outer dia. 26 ±1.0 M.M
2 Facing Turning Facing length 87.7 ±0.7 V.C
lathe
3 Rough Turning Diameter 24 ±0.2
turning lathe Length 87.7 ±0.7
4 Collar side Turning Diameter 18 ±0.5 S.G
turning lathe
Length 21 ±0.4 S.G
5 Bolt side Turning Diameter 12 ±0.3 S.G
turning lathe Length 39 ±1.0 V.C
Chamfering 1.5x45 M.M
6 Thread Tapping Thread length 16.5 ±0.5 S.G
rough machine
7 Final thread Tapping Threads M12 X 1.25 T.R.G
machine
Dimension 39 ±1.0 S.G
Dimension 27.7 ±0.4 L.G
8 Inspection Gauges/ Dimension 21 ±0.4 V.C
Instrument Diameter 18 ±0.5 S.G
s Diameter 12 ±0.3 S.G
Chamfer 1.5x45 V.C
Thread M12 X 1.25 T.R.G
Thread length 16.5 ±0.5 S.G
Appearance No scratch no visual
thread damage

www.smenec.org 49 © SME
 Journal of Manufacturing Engineering, March 2018, Vol. 13, Issue. 1, pp 042-050

Fig. 14 Overall % defectives

3.5 Control Phase Fig. 15 Process capability results

Having done the hard work, it is time to get the
control initiated so that the process does not go back to References
its old state. If the Six Sigma changes and procedures
are not maintained, it will lose its performance. The 1. Chaudhari P P (2012), “Key for implementation of Six
Sigma in Indian industries”, Journal of Information and
major areas to be included in the control phase are Operations Management, Vol.3(1), 203– 205.
Planning, Documentation, Process controls, Monitoring
and System review. 2. Arumugam V, Antony J and Kumar M (2013), “Linking
learning and knowledge creation to project success in Six
Sigma projects: An empirical investigation. Journal of Int.
3.5.1 Standard Operating Procedure (SOP) J. Production Economics”, Vol.141, 388–402.
Standard Operating Procedure is very
3. Kuptasthien N and Boonsompong T (2011), “Reduction of
descriptive and gives clear instruction to the operator Tombstone Capacitor Problem by Six Sigma Technique: A
what needs to be done, what precautions to be taken , Case Study of Printed Circuit Cable Assembly Line”,
what data to be prepared with great care and how to Thailand IEEE 546-550.
report in case of any problems. The standard operating 4. Plecko A and Polajnar A (2009), “An application of Six
procedure for the improved results are given table 5. Sigma in manufacturing company”, Advances in
production and management, Vol. 4(4), 243-254.
5. Hongbo Wang (2009), “A Review of Six Sigma Approach:
4. Results and Conclusion Methodology”, Implementation and Future Research,
IEEE 1-4.
The process Sigma level through Six Sigma
6. Sahoo A K, Tiwari M K (2007), “Six Sigma based
DMAIC methodology was found to be approaching 4.11 approach to optimize radial forging operation variables”,
Sigma from 2.67 Sigma, while the process yield Journal of Materials processing technology, Vol. 202, 125-
increased from 87.8% to 99.6% This Six Sigma 136.
improvement methodology, DMAIC project shows that 7. Soković M, Pavletić D and Krulčić E (2006), “Six Sigma
the performance of the company is increased to a better process improvements in automotive parts production”,
level to enhancement in customer’s satisfaction, Journal of achievements in material and manufacturing
engineering, vol.19(1) 96-102.
conformity of delivery schedules, development of
specific methods to redesign and reorganize a process 8. Sokovic M, Pavletic D and Fakin S (2005), “Application
with a view to reduce errors and defects. of Six Sigma methodology for process design”, Journal of
Materials Processing Technology, Vol. 162–163, 777–783.
1. Sigma impact: The Sigma level has been increased
from 2.67 Sigma (previous process PPM 120000) to 9. Sokovic M, Pavletic D and Fakin S (2005), “Application
of Six Sigma methodology for process design quality
4.11. improvement. Journal of achievements in material and
2. Improvement in process yield: The process yield is manufacturing engineering”, Vol.19(1), 611-614.
improved by optimum utilization of resources from
10. Mikel Harry, web resource:www.isixsigma.com
87.8% to 99.6%.
3. COPQ impact: COPQ of Cushion P-70 Bolt RR is 11. Greg brue 2006, Six Sigm for small business, Tata Mcgraw
reduced from 25% to 12% per year. hill-208 pages
4. Process Capability increased to 1.93and is presented 12. Gopalakrishnan N, (2010), Simplified lean manufacturing
in Fig.15. elements, rules,tools and economy, PHI learning pvt. ltd.,
new Delhi.

www.smenec.org 50 © SME