Analyze Phase in Lean Six Sigma: Understanding Process Behavior and Validating Root Causes

What Is the Analyze Phase?

The Analyze phase of DMAIC focuses on understanding why a problem is occurring by examining data, identifying patterns, and validating potential causes. Rather than relying on assumptions or intuition, this phase uses both Lean and Six Sigma methods to:

Explore relationships within the data
Identify sources of variation and inefficiency
Confirm which factors are truly driving the problem

The objective is verify the root causes of problems, waste, and process constraints that impede the flow of value creation and achieving target service and product capability.

Why the Analyze Phase Matters

The Analyze phase ensures that improvement efforts are focused, effective, and evidence-based. Without proper analysis:

Solutions may address symptoms rather than causes
Time and resources may be wasted on low-impact changes
Problems are likely to persist or reoccur

This phase provides the clarity needed to answer:

What is really causing the problem?
Which factors have the greatest impact on performance?
Where should we focus improvement efforts?

Key benefits include:

Reducing risk of incorrect solutions
Increasing confidence in decision-making
Prioritizing high-impact opportunities
Bridging data from Measure into actionable insight

Start Learning for Free

Key Concepts in the Analyze Phase

🔹Root Cause Identification and Validation

Identifying and confirming root causes using data to confirm their true impact.

🔹Process Behavior and Variation

Understanding how a process trends, variation, patterns, and stability over time.

🔹Data Relationships and Correlation

Exploring how different variables interact and influence outcomes.

🔹Current State Value Stream Mapping

Complete value stream mapping by acquiring relevant process data to identify waste and constraints to flow.

🔹Statistical Analysis and Hypothesis Testing

Using data to scientifically test assumptions and assess overall current state process/system behaviour.

Common Tools Used in the Analyze Phase

A Note on Tool Selection: While the following tools are commonly used in the Analyze phase, not every tool is required for every project. The selection of tools depends on the nature of the problem, the process being studied, and the type of data available. Effective Continuous Improvement practitioners focus not on using more tools—but on using the right tools at the right time. The goal is not complexity, but clarity and confidence in overall understanding of behaviour.

🔹Value Stream Mapping - Current State

Complete current state value stream map including relevant process data and identify opportunities for improvement.

🔹Failure Mode and Effects Analysis

Complete FMEA with analysis of severity, occurrence, and detection assessment and resulting risk priority numbers.

🔹Statistical and Graphical Analysis

Continue to use the tools employed in the "Measure" phase as new data is gathered from the process.

🔹Process Capability Studies

Update and confirm short and long-term process capability for discrete and continuous variables (data).

🔹Hypothesis Testing

Employ scientific methods to confirm assumptions about the process using data sampled from a process.

🔹Statistical Analysis and Modeling

Use of statistical tools including regression analysis, analysis of variance (ANOVA), general linear model (GLM), and logistic regression to model process behaviour.

Beyond the Basics: Advanced Statistical Analysis

Lean Six Sigma analysis extends well beyond foundational tools. As problems become more complex, deeper statistical methods are often required to fully understand process behavior and validate results with confidence. These advanced methods build on a foundation of statistical thinking and may include:

Probability distributions and statistical inference
Confidence intervals and hypothesis testing
Correlation and regression analysis
Analysis of variance and comparison of multiple groups
Data transformation and non-normal data analysis
Predictive modeling and reliability analysis

At RPM-Academy, these concepts are introduced progressively:

Yellow Belt – foundational awareness and basic interpretation
Green Belt – practical application and intermediate analysis
Black Belt – advanced statistical modeling and data-driven decision making

👉 The goal is not to master every technique at once—but to build the knowledge and confidence to apply the right level of analysis for each situation. Browse our online course catalog - select the "Lean Six Sigma" filter.

BROWSE COURSE CATALOG