DMAIC Problem Solving Guide

A comprehensive reference guide for the Six Sigma DMAIC methodology

This guide provides key points and coaching questions for each phase: Define, Measure, Analyze, Improve, and Control. Use it as a reference for data-driven problem solving projects.

Six Sigma DMAIC - Data-Driven Problem Solving

DEFINE / MEASURE / ANALYZE

Define

Define the problem, project goals, and customer requirements

Key Points

•Problem Statement: Clearly articulate the issue, its impact, and why it matters to customers and business
•Project Scope: Define boundaries - what's included and excluded from this DMAIC project
•Voice of Customer (VOC): Identify customer requirements and expectations (CTQs - Critical to Quality)
•Business Case: Quantify financial impact, resources needed, and expected benefits
•Project Charter: Document team members, timeline, goals, metrics, and stakeholder buy-in
•High-Level Process Map: Create SIPOC (Suppliers, Inputs, Process, Outputs, Customers) diagram

Coaching Questions

→Who is the customer and what are their critical requirements?
→What is the business impact if this problem remains unsolved?
→Do we have clear project boundaries and executive sponsorship?
→Is the problem statement focused on the 'what' rather than the 'why' or 'how'?

Measure

Measure current process performance and establish baseline data

Key Points

•Define Metrics: Identify Y (output) and X (input) variables to measure process performance
•Data Collection Plan: Determine what data to collect, how, when, and who will collect it
•Measurement System Analysis (MSA): Validate that measurement methods are accurate and repeatable
•Process Capability: Calculate baseline performance using Cp, Cpk, Sigma level, or defect rates
•Detailed Process Map: Document current state with value stream or detailed flow diagrams
•Establish Baseline: Capture current performance data to enable before/after comparison

Coaching Questions

→Are we measuring outputs (Y) that matter to customers and inputs (X) that drive those outputs?
→Is our measurement system reliable and repeatable (MSA validated)?
→What is the current process capability and sigma level?
→Do we have sufficient historical data or do we need to start fresh data collection?

Analyze

Analyze data statistically to identify and verify root causes

Key Points

•Statistical Analysis: Use graphical tools (histograms, box plots, scatter plots, Pareto charts) to visualize data patterns
•Variation Analysis: Identify sources of variation using control charts, multi-vari analysis, and capability studies
•Correlation & Regression: Determine relationships between input variables (Xs) and output variables (Ys)
•Hypothesis Testing: Validate suspected causes statistically using t-tests, chi-square, ANOVA, or regression analysis
•Root Cause Verification: Test each suspected cause with data and confirm with statistical significance (p-values, confidence intervals)
•Prioritize Vital Few: Focus on statistically significant root causes using Pareto principle and effect size analysis

Coaching Questions

→What does the statistical analysis reveal about root causes vs. symptoms?
→Have we validated suspected causes with hypothesis tests and confirmed statistical significance?
→Which input variables (Xs) have the strongest statistical relationship to the output (Y)?
→What is the magnitude of variation from each root cause (common vs. special cause)?

IMPROVE / CONTROL

Improve

Develop, test, and implement solutions to eliminate root causes

Key Points

•Generate Solutions: Brainstorm countermeasures that address validated root causes
•Solution Selection: Evaluate options using criteria like impact, feasibility, cost, and risk
•Pilot Testing: Test solutions on small scale before full implementation (PDCA cycles)
•Design of Experiments (DOE): Optimize multiple variables simultaneously if applicable
•Mistake-Proofing (Poka-Yoke): Design solutions to prevent errors from occurring
•Implementation Plan: Define who, what, when, where, and how solutions will be deployed
•Change Management: Prepare people, training, and communication for new processes

Coaching Questions

→Do proposed solutions directly address the root causes we validated?
→Have we piloted solutions and measured results before full rollout?
→What might go wrong during implementation and how will we mitigate risks?
→Are people trained and ready to adopt the new process?

Control

Control and sustain improvements over time

Key Points

•Control Plan: Document how the improved process will be monitored and maintained
•Statistical Process Control (SPC): Implement control charts to detect shifts or trends
•Standard Operating Procedures: Update work instructions, visual aids, and documentation
•Training & Handoff: Train process owners and operators on new standards
•Monitoring Dashboard: Create visual management system for key metrics (Y and X variables)
•Response Plan: Define actions to take when process goes out of control
•Continuous Improvement: Schedule periodic reviews and identify next improvement opportunities
•Project Closeout: Document results, financial benefits, and lessons learned

Coaching Questions

→How will we know if the process starts to drift back to old performance?
→Who owns this process and will sustain the improvements long-term?
→Are control limits and reaction plans clearly defined and communicated?
→What are the next improvement opportunities for this process?