Course Content
Define
Identify the problem or process that needs improvement, set project goals, and outline the scope.
0/10
Six Sigma Management at Corporate Level
Six Sigma Project Team Structure
Create team roles and responsibilities
Create a Problem Statement
Create the SIPOC process map
What is CTQ (Critical to Quality)
Create CTQ Tree
Creating a Project Charter
Stakeholder Analysis
Outcome expected after the Define Stage
Measure
Collect data to establish baselines, understand current performance, and quantify the problem. For example, measuring the average turnaround time for policy renewals.
0/4
FMEA (Failure Modes and Effect Analysis) concept
Data collection methods
Introduction to data sampling
Use of Run chart for analysis
Analyze
Examine the data to identify root causes of inefficiencies or risks. This might involve analyzing claims data to uncover bottlenecks or sources of errors.
0/12
5 Whys
Root cause analysis with Fishbone Diagram
Lean 7 Wastes concept
Value chain mapping with Value Stream map
What is a process
Components of process
Process Mapping with Swimlane diagram
Process redesign principles
80/20 analysis with Pareto Diagram
Statistical analysis with Box Plot
Hypothesis Testing
Correlation and Regression Analysis
Improve
Develop and implement solutions to address root causes. For example, streamlining workflows or introducing new digital tools to reduce manual errors.
0/4
Solution Selection Matrix
Cost Benefit Analysis
Risk Assessment
Implementation Planning
Control
Put controls in place to sustain improvements, such as regular monitoring, updated procedures, or dashboards for ongoing performance tracking.
0/4
Create Control Plan
Create Key Performance Indicator (KPI)
Prepare the Process and User Guides
Handover to Process Owner
Six Sigma DMAIC (Define, Measure, Analyze, Improve, Control)

What is Hypothesis Testing

Hypothesis Testing allows Six Sigma team to draw conclusions about the population based on statistical analysis performed on a sample. Because the conclusions are based on samples and not the entire population, there is always some risk of error.

Using the information about the sample, statistically the p-value of the data collected can be calculated. p-value is the mean to tell if the assumption or conclusion drawn on the sample is correct or not.

 

Normal Probability Distribution

Normal Distribution, also called Gaussian distribution, is probably the most important distribution related to continuous data from statistical analysis standpoint.

Since many natural process outcomes are normally distributed, Six Sigma relies on this curve and the Empirical Rule (68-95-99.7%) to assess how well a process meets specifications and whether a sample size is good enough to be used for analysis.

In a perfect normal distribution, 68.26% of all data points fall within plus or minus 1 standard deviation from the mean; 95.46% of the data points fall within plus or minus 2 standard deviation from the mean; and 99.73% of the data points fall within plus or minus 3 standard deviation from the mean.

Testing whether data is normal is critical in statistical analysis because the results of tests conducted can be invalid if the accuracy of the data is not accounted for.

To determine if the sample data used is normal, advanced statistics are required. Microsoft Excel and other programs can be used to perform the calculation before the data is to conduct required analysis.

 

How to use Excel to check if sample data is normal

1. Create a Histogram using the sample data. Using the example of Time taken for Adults to complete the 100m Swim, the following Histogram is created.

Six Sigma Histogram in Excel

 

2. The Descriptive Statistical information of the data set is created as followed

Mean 155.4117647
Standard Error 14.14578996
Median 134
Mode #N/A
Standard Deviation 58.32458618
Sample Variance 3401.757353
Kurtosis 0.815962627
Skewness 0.990489892
Range 220
Minimum 80
Maximum 300
Sum 2642
Count 17

 

3. The Expected Distribution and the Observed Distribution are calculated as followed

Bin Range CDF to left Bin Only Expected Number Observed (Exp – Obs)2 Divided by Exp
70 0.07153944 0.07153944 1.216170474 0 1.47907062 1.216170474
80 0.098011238 0.026471798 0.45002057 1 0.30247737 0.672141216
90 0.131034842 0.033023604 0.561401273 1 0.19236884 0.342658366
100 0.171041282 0.04000644 0.680109482 1 0.10232994 0.150460986
110 0.218106468 0.047065185 0.800108148 1 0.03995675 0.049939189
120 0.27187573 0.053769262 0.914077456 1 0.00738268 0.00807665
130 0.331528805 0.059653075 1.014102274 3 3.94378978 3.888946783
140 0.395796993 0.064268188 1.0925592 1 0.00856721 0.007841411
150 0.46303638 0.067239388 1.143069588 1 0.02046891 0.017906965
160 0.531351356 0.068314975 1.161354579 0 1.34874446 1.161354579
170 0.59875332 0.067401965 1.145833402 1 0.02126738 0.018560622
180 0.663332671 0.064579351 1.097848959 0 1.20527234 1.097848959
190 0.723419496 0.060086825 1.021476033 1 0.00046122 0.000451523
200 0.77771068 0.054291184 0.922950123 2 1.16003644 1.256878794
210 0.825347616 0.047636936 0.809827904 1 0.03616543 0.044658162
220 0.865937963 0.040590347 0.690035899 0 0.47614954 0.690035899
230 0.89952457 0.033586608 0.570972328 0 0.3260094 0.570972328
240 0.926512771 0.0269882 0.458799406 1 0.29289808 0.63840118
250 0.947572176 0.021059405 0.35800989 0 0.12817108 0.35800989
260 0.96353033 0.015958154 0.271288615 0 0.07359751 0.271288615
270 0.975273451 0.011743121 0.199633057 0 0.03985336 0.199633057
280 0.983665127 0.008391677 0.142658502 0 0.02035145 0.142658502
290 0.989488549 0.005823421 0.098998165 0 0.00980064 0.098998165
300 0.993412937 0.003924388 0.066714599 1 0.87102164 13.05593757

 

4. Using the result above, the p-Value can be calculated using the Chi-Squared formula in Excel, and it is derived as 0.207984876. And it is good since the p-Value is greater than 0.05

Total x2 25.95982989
Degree of Freedom 21
CHISQ.DIST 0.207984876
Previous
Next