Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 13 – Measure Phase (DMAIC) - Coggle Diagram
Chapter 13 – Measure Phase (DMAIC)
Purpose of the Measure Phase
Move from assumptions to facts using reliable data
Understand current process performance before making changes
Establish a baseline to compare future improvements
Key Performance Metrics
Determine process capability early if data is available
Calculate sigma level to understand defect performance
Ensure metrics reflect customer and business impact
Use metrics like DPMO, First Time Yield (FTY), and Rolled Throughput Yield (RTY)
Failure Modes and Effects Analysis (FMEA)
Identify where and how a process can fail
Prioritize risks before deep analysis begins
Bridge between Define and Analyze phases
Focus improvement efforts on highest-risk failures
Data Collection Foundations
Build a clear baseline of current performance
Use historical data when available
Collect new data when gaps exist
Ensure data is accurate before analysis begins
Core FMEA Elements
Process steps and possible failure points
Effects of failure on the customer
Severity, occurrence, and detection ratings
Risk Priority Number (RPN = SEV × OCC × DET)
Types of Data
Discrete (Attribute) Data
Categorical or descriptive information
Examples include pass/fail or yes/no outcomes
Displayed using Pareto charts, bar charts, or pie charts
Easier to collect but less detailed
Continuous (Variable) Data
Numerical data measured on a scale
Examples include time, temperature, or length
Displayed using histograms, box plots, and run charts
Provides more precision and insight
Choosing the Right Data Type
Prefer continuous data when possible
Continuous data can be converted to discrete if needed
Discrete data cannot be converted back to continuous
Using FMEA Effectively
Complete risk scoring during Measure phase
Plan corrective actions during Improve phase
Recalculate RPN after changes to confirm risk reduction
Levels of Data Measurement
Nominal: Categories with no ranking (e.g., state names)
Ordinal: Ranked categories with no fixed spacing (e.g., ratings)
Interval: Ordered values with meaningful differences but no true zero
Ratio: Highest level with a true zero and full analytical capability
Gage R&R Studies
Attribute Gage R&R
Used for pass/fail or yes/no data
Measures agreement between appraisers
Identifies subjectivity and inconsistency
Variable Gage R&R
Used for continuous measurements
Analyzes variation caused by operators and tools
Evaluated using statistical thresholds and categories
Sampling Strategies
Use sampling when full population data is not available
Avoid Non-Random Sampling
Convenience sampling introduces bias
Judgment sampling skews results toward best cases
Baseline Metrics & Visualization
Baseline metrics define current performance
Visual displays improve understanding and decision-making
Run Charts
Show process performance over time
Highlight trends, shifts, and patterns
Often include a median line for reference
Useful for monitoring before control charts
Measure Phase Tollgate Checklist
Key metrics are clearly defined and agreed upon
Measurement systems are validated and reliable
Sampling method supports valid conclusions
Baseline performance is documented and visualized
Sponsor approves readiness to move to Analyze phase
Common Sampling Methods
Sequential sampling for time-based or interval data
Stratified sampling to reduce subgroup bias
Simple random sampling for equal selection chance