Day 1 : Completed

LMS Tour

Introduction

Project

Simulation Test

Certification

Quality : Is an intangible characteristic of any product or service.

Juran :

a) Fitness for use

b) Give so much to the customer which he values : not more

Push Theory : Producers need to produce their products & push them into market

Quality is customer satisfaction aimed at meeting :

a) Stated Needs : Customer would talk about

b) Unstated Needs : Customer will not talk about but will not buy the product if it’s not there in
the product or service
c) Delighters : Little extra to buy the market share

Kano Model :

Motorola : If a product or process or service is producing less than or equal to 3.4 defects per million
opportunities, that product or process or service is performing at 6 Sigma level

Defect : Something not meeting the set standards

Six Sigma Methodology :

For Existing Product/Process/Service

Define (Identify a real problem) ---Measure (Statistical Problem)--- Analyse (Statistical Solution)
---- Improve (Real Solution & Check) ---- Control (Implement the real solution)
For New Product/Process/Service : DFSS : Design For Six Sigma :

1. DMADV : New product/process/service for the business but not a new concept for the world
2. IDOV : Brand new concept of product/process/service for both business & world

Customers : 2 Types :

Internal External

Employees Stockiest

Managers Dealers/ Resellers

Departments End Users / Buyers

Voice Of the Customer : VOC : 2 methods :

Proactive Reactive

Survey Feedbacks

Focus Group Complaints

Interview Benchmarking : Internal or External

Day 2 : Completed

VOC - VOP (Voice Of Process) --CTQ/CTP/CTD/CTS (Critical To Quality/Price/Delivery/Safety)

Define Phase :

Project charter : 5 Parts

a) Problem Statement : Crisp, data driven pin point the exact pain area
b) Goal Statement : SMART
c) Business Case : Benefits of doing the project & the consequences of not doing the project
d) Project Scoping : What are within the bounds of the project & what are the out of bound
situations
e) Milestones : Timelines

Gantt Chart :
CPM : Critical Path Method : The longest path in the project with no float

Day 3 : Completed

PERT : Project Evaluation & Review Technique :

Kick Off Meeting : The official start of the Six Sigma Project. Process Owner & Six Sigma Lead will be
surely present in the meeting.

Team : 4-7 members

Team Dynamics

Group Challenges

RACI : Responsible Accountable Consult Inform

Affinity Diagram :

Interrelationship Diagram :
Tree Diagram

Lean Sigma : Waste Reduction

Muda : Wastes

Defect : Something not meeting the set standards

Rework : Fix the defects

Pull : Customer orders & then the entire production cycle goes through

JIT : Just In Time

Day 4 : Completed

Quality Tests : 2 types :

1. Destructive Tests : DT : Product gets chemically & or physically damaged after the test. Can’t
be tested on every unit & hence sample test are done. Eg : Car crash test, chain pull test,
yarn strength test, nuclear weapon test etc.
2. Non Destructive Tests : NDT : Product doesn’t get chemically & or physically damaged after
the test. Tests can be done on every single unit. Eg. X – ray, caret meter test, ultrasonic test
etc.

3M:
Value Stream Mapping :

Shopfloor : Gemba

From the eyes of the customer :

1) VA : Value Adds : The customer is willing to wait for & or pay for
2) NVA : Non Value Add : The customer is not willing to wait for & or pay for
3) RNVA : Required Non Value Add : The NVAs which are required by the process. Eg. Quality
check

5S :

Kanban :
Kaizen :

Kaizen Blitz follow :

Poka Yoke : Mistake Proofing :

RCO : Reduce Chances of Occurrence

ICD : Increase Chances of Detection

Takt Time

Jidoka

Day 5 : Completed

SIPOC : Supplier Input Process Output Customer

Fever Causes for fever

Wet in the rain

Ate an ice cream

Contaminated

Output Inputs

Monitor Control

Dependent Independent

Y Xs

Y = f(Xs)

Process : A series of activities with a definite start & end step one activity leading to the other
converting some inputs to outputs

High Level Process Mapping : Define Phase

Detailed Process Mapping : Measure Phase

Flowchart :

FMEA : Failure modes & effects analysis

RPN : Risk Priority Number

RPN = S *O * D

S : Severity

O : Occurrence

D : Detection

RPN > 80---Process reengineering

QFD : Quality Function Deployment

HOQ : House of Quality

Toll Gate Review : After every phase the team meets for toll gate review

Day 6 : Completed

Measure Phase :

Defect Opportunity : O : Number of chances of creating defects while processing one unit
O=3

Total Number of units processed (N) = 10

Total Defects : D = (3 + 2 + 1) = 6

Defect Per Opportunity : DPO = D/(N * O) = 6/(10 * 3) = 6/30 = 1/5 = 0.2

Defects Per Million Opportunities : DPMO = DPO * 1000000 = 0.2 * 1000000 = 200000

Yield : Defect Free Output; yield is expressed in percent

Yield = (1 – DPO) * 100 = (1 – 0.2) * 100 = 0.8 * 100 = 80%

Look in the sigma table the corresponding Sigma Level

Defective Unit : Unit that has at least one defect but may have more than one defect

2, 10, 6, 4, 8, 100 : 2, 4, 6, 8, 10, 100

1. Measures of Central Tendency :

a) Mean : Average (X̅) : Sum of all/N = (2 + 10 + 6 + 4 + 8)/5 = 30/5 = 6; mean gets affected
by outliers
b) Median : Exact centre of the data set : Arrange the data set in ascending order, For odd
data count : ((n + 1)/2}th Term = {(5 + 1)/2}th Term = (6/2)th term = 3rd term, for even
data count [(n/2)th Term + {(n/2) + 1}th Term]/2 = = [(6/2)th Term + {(6/2) + 1}th
term]/2 =[3rd term + 4th Term]/2 = [6 + 8]/2 = 14/2 = 7; Median doesn’t get affected by
outliers

Xi = 2, 4, 6, 2, 6, 2, 2, 4, 8, 2, 6, 6, 2, 8, 4, 2

c) Mode : The X-Value which has the highest frequency; frequency is the number of times a
particular data point repeats itself in the data set; to have mode we need fluctuations in
the frequency of the data points

Xi f CF
2 7 7
4 3 7 + 3 = 10
6 4 7 + 3 + 4 = 14
 8 2 7 + 3 + 4 + 2 = 16

Outliers : Extreme Values with respect to the rest of the data set

2. Measures of Dispersion :
a) Range : Maximum Value – Minimum Value = 10 – 2 = 8; range gets immensely affected by
outliers
b) Variance :
c) Standard Deviation : Variance ^ 0.5

Xi X̅ (Xi - X̅) (Xi - X̅)^2

2 6 (2 – 6) = -4 16

4 6 (4 – 6) = -2 4

6 6 (6 – 6) = 0 0

8 6 (8 – 6) = 2 4

10 6 (10 – 6) = 4 16

∑(Xi - X̅)^2 = 40

VarianceSample : {∑(Xi - X̅)^2}/(n – 1) = 40/(5 – 1) = 40/4 = 10

St. Dev.Sample = Variance ^ 0.5 = 10 ^ 0.5 = 3.16

VariancePopulation = {∑(Xi – X-Bar)^2}/(n) = 40/(5) = 40/5 = 8

St. Dev.Population = Variance ^ 0.5 = 8 ^ 0.5 = 2.82

Data :

1. Population Data : Entire data gamut

2. Sample Data : Subset of the population data

Day 7 : Completed

For the first time when we are setting up the process limits :

Target = 10 cms

Process Mean = 10.9 cms

St. Dev. = 0.3

LSL (Lower Specification Limit) = Actual process mean – (3 * St.Dev) = 10.9 – (3 * 0.3) = 10.9 – 0.9

= 10 cms

USL (Upper Specification Limit) = Actual process mean + (3 * St.Dev) = 10.9 + (3 * 0.3) = 10.9 + 0.9
 = 11.8 cms

Unpredictable, Off-Target & On-Target

Process Capability :

Cp & Cpk

Cp : Instantaneous Process Capability

Cpk : Short Term Process Spread

Cp = (USL – LSL)/(6*St. Dev.ST)

Cpk = min (Cpl, Cpu)

Cpl = (X̅ – LSL)/3*St.DevST

Cpu = (USL – X̅)/3*St.DevST

Ideal Situation:

Cp = Cpk = 2 --- Process is at 6 Sigma level & aligned to the target

Marginal Capability :

Cp = Cpk = 1 --- Process is at 3 Sigma level & aligned to the target

Process Indices :

Pp & Ppk

Pp = (USL – LSL)/(6 * St. DevLT)

Ppk = Min {Ppl, Ppu}

Ppl = (X̅ - LSL)/(3 * St. DevLT)

Ppu = (USL - X̅)/(3 * St. DevLT)

Measurement System Analysis (MSA) :

Total Variation = Actual Process Variation + Measurement System Variation

Accuracy : To know about accuracy we need to know the true value

Precision : Closeness of the data points

Gage R & R Test :

Repeatability : One operator measuring one unit using one gage multiple times

Reproducibility : Multiple operators measuring one unit using one gage

We Check :

EV (Equipment Variation) : Variation due to equipment

AV: (Appraiser Variation) : Variation due to the appraiser

Gage Capability : Overall gage performance

Generally :

If the variation is < 10%, it is ok

If the variation is between 10% to 30%, it may be acceptable

If the variation is > 30%, it is not acceptable

Bias : Bias examines the difference between the observed average measurement and a reference
value. Bias indicates how accurate the gage is when compared to a reference value.

Linearity : Linearity examines how accurate your measurements are through the expected range of
the measurements. Linearity indicates whether the gage has the same accuracy across all reference
values.

Stability : One operator measuring one unit using one gage over a prolonged time frame.

Day 8 : Completed

Data : 2 Types :

a) Variable Data or Continuous Data or Quantitative Data : Data that can be measured directly
in universally established & well accepted scales. This data can take any value on the
number line. Eg. Weight, height, speed, time etc. This data follows normal distribution.
b) Attribute Data or Discrete Data or Qualitative Data : Data that cannot be measured directly
in universally established & well accepted scales but need to internally design a scale to
capture such data. This data takes whole numbers. The counts are Poisson data. Eg.
Employee id, shoe sizes, T-Shirt sizes etc. The data for which we get only two outcomes are
binomial data. Eg. Off/on, yes/no, go/no go, pass/fail, defect/no defect etc.

With variable data we use Gage R & R worksheet

With attribute data we use Attribute worksheet

Generally :

Effectiveness
Acceptable >= 90%
Marginal >=80%
Unacceptable <80%

7 tools of quality :

1. Flowchart : Pictorial representation

2. Check Sheet : Tally Marks
3. Pareto Diagram : We should work on vital few (1 or 2) 20% of the problems & bring in 80%
overall improvement

4. Fishbone Diagram/Ishikawa/5M&1E/6M/RCA/5 Why :

5. Histogram : Frequency Plot Chart :

6. Scatter Plot :
 7. Control Charts :

Stem & Leaf plot

Box Plot

Analyse Phase :

Hypothesis Testing

Null Hypothesis : Ho or ho : The groups are similar

Alternate Hypothesis : Ha or ha : The groups are dissimilar

We perform the tests as per the conditions :

a) If we have 2 groups of data with data points less than or equal to 30 in each group & we
want to compare the means, we perform T-Test
b) If we have 2 groups of data with data points more than 30 in each group & we want to
compare the means, we perform Z-Test
c) If we have 2 groups of data & we want to compare the variance, we perform F-Test
d) If we have 3 or more groups of data, we perform ANOVA (Analysis Of Variance)
e) If we have one group of data & we want to compare it with a target; we perform Chi-
squared test

We will check the p-value : P-value is the probability of accepting or rejecting the null hypothesis

1. If the p-value ≤ 0.05, ---- reject Null Hypothesis

2. If the p-value > 0.05, ---- accept null hypothesis

Probability : chances of occurrence, we can express in percent or fraction or decimal