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)

Why Data needs to be collected

The key output of the Measure phase is to have a good understanding of the current process performance level with the figures such as Sigma level and some other process-specific metrics

 

Types of Data

Data can be grouped into discrete data or continuous data:

Discrete Data

Discrete data is categorical in nature. It falls into 3 categories: ordinal, nominal or binary. 

Ordinal Data – The numbers/symbols are qualitative in nature, but they are also ranked. Central tendencies with ordinal data are measured by either the mode or the median. For example, student test scores can be expressed in ordinal fashion via grades A, B, C, D and F

Nominal Data – The numbers/symbols are assigned to each category but they don’t provide any information if the data is better or worst than other data in the listing. For example, 1. is assigned to “product that is produced in Thailand” for a company, and 2. is assigned to “product that is produced in Malaysia”; the no, of 1s and 2s produced by the company does not tell if one of them is better than the other.

Binary Data – The numbers/symbols assigned to the data has only 2 states. In the student test results example, a Pass/ Fail can be assigned to each student’s result

Note: Discrete data are best displayed via Pareto chart, Pie chart and Bar Chart

Continuous Data

Continuous data is quantitative data and is measured in units. For example, the time of day is measured in hours.

Note: 

  • Almost all continuous data can be converted into percentage
  • Continuous data is best visualized in graph using Histogram and box plots.

 

Choosing between Discrete and Continuous data

Discrete or Continuous data may be chosen depending on the purpose of measurements. In the nutshell, Discrete data is easier to collect, but it provides less information compare to the Continuous data.

The Continuous data is typically more time-consuming to collect than Discrete data unless teams have access to automated or computerized data collection. 

Where possible Continuous data would be preferred over Discrete data as:

  • It provides more information than discrete data
  • It is more precise than discrete data
  • It reduces the variation and errors inherent in estimation and rounding (that happens in some Discrete data)

 

Previous
Next