Description

This curriculum spans the breadth of a multi-workshop Six Sigma deployment, addressing data management tasks typically encountered in end-to-end process improvement initiatives, from defining CTQs and validating measurements to sustaining control plans and integrating data across projects in regulated environments.

Module 1: Defining Data Requirements in the Define Phase

Selecting critical-to-quality (CTQ) metrics based on customer specifications and operational feasibility
Mapping data sources to process boundaries using SIPOC diagrams to identify measurable inputs and outputs
Aligning data collection objectives with project charters and stakeholder expectations
Deciding whether to use existing enterprise data or initiate new data capture systems
Establishing data ownership roles across departments to prevent duplication and gaps
Designing preliminary data collection plans with clear operational definitions for each metric
Assessing data accessibility constraints due to legacy systems or IT policies
Documenting data privacy requirements when handling customer or employee information

Module 2: Data Collection Planning in the Measure Phase

Choosing between continuous and discrete data based on measurement system capabilities and analysis needs
Designing sampling strategies that balance statistical power with operational disruption
Developing standardized data collection forms with built-in validation rules
Conducting pilot data collection runs to identify process interruptions or measurement errors
Calibrating measurement tools and documenting calibration schedules for audit compliance
Training data collectors on consistent interpretation of operational definitions
Implementing timestamping and version control for collected datasets
Integrating real-time data feeds from manufacturing or transactional systems when available

Module 3: Measurement System Analysis (MSA)

Selecting appropriate MSA method (e.g., Gage R&R, attribute agreement analysis) based on data type
Randomizing measurement order to prevent temporal bias in operator assessments
Quantifying repeatability and reproducibility thresholds acceptable for the process under study
Handling destructive testing scenarios where repeated measurements are not possible
Interpreting variance components to determine if measurement error exceeds process variation
Requiring recalibration or retraining when MSA results exceed predefined tolerance levels
Documenting MSA results for regulatory compliance in audited environments
Updating measurement protocols based on MSA findings before full-scale data collection

Module 4: Data Validation and Cleaning Procedures

Implementing automated outlier detection using control charts or z-score thresholds
Establishing rules for handling missing data (e.g., imputation, exclusion, flagging)
Validating data entry against predefined ranges and logical constraints
Resolving discrepancies between multiple data sources reporting the same metric
Tracking data lineage from source systems to analysis datasets
Creating audit logs for all data transformations and cleaning operations
Verifying timestamp accuracy across distributed systems for time-series analysis
Standardizing units of measure and categorical labels across departments

Module 5: Statistical Process Control and Baseline Performance

Selecting appropriate control charts (e.g., I-MR, X-bar R, p-chart) based on data type and subgroup size
Calculating process capability indices (Cp, Cpk) using validated baseline data
Distinguishing between common cause and special cause variation using run rules
Handling non-normal data through transformation or non-parametric methods
Setting control limits based on historical performance while accounting for recent process changes
Updating baselines after process shifts are confirmed and documented
Integrating control charts into operational dashboards with escalation protocols
Defining response procedures for out-of-control signals in real-time monitoring

Module 6: Data Analysis in the Analyze Phase

Choosing hypothesis tests (t-tests, ANOVA, chi-square) based on data distribution and sample size
Validating assumptions of statistical tests before interpreting results
Using regression analysis to isolate significant input variables affecting output performance
Applying Pareto analysis to prioritize root causes based on impact magnitude
Interpreting interaction effects in multifactorial experiments
Managing false discovery rates when conducting multiple simultaneous comparisons
Documenting analytical decisions, including rationale for test selection and data exclusions
Presenting statistical findings using visualizations that prevent misinterpretation

Module 7: Data-Driven Decision Making in the Improve Phase

Designing pilot implementations with control and test groups for valid comparison
Specifying success criteria in measurable terms before launching improvement interventions
Collecting pre- and post-intervention data using consistent measurement protocols
Using design of experiments (DOE) to optimize process settings with minimal trial runs
Assessing scalability of improvements based on pilot data and resource constraints
Updating data collection systems to reflect new process configurations
Conducting risk assessments on proposed changes using failure mode and effects analysis (FMEA)
Securing cross-functional sign-off on data-based recommendations before full rollout

Module 8: Control Plan Development and Sustainment

Transferring ownership of control charts and dashboards to process operators
Embedding data monitoring responsibilities into standard operating procedures (SOPs)
Defining response plans for out-of-control conditions with escalation paths
Setting frequency and method for ongoing measurement system revalidation
Integrating control plan documentation into enterprise quality management systems
Scheduling periodic audits of data integrity and compliance with control protocols
Archiving project data according to retention policies and legal requirements
Establishing feedback loops for continuous data quality improvement

Module 9: Governance and Cross-Project Data Integration

Aligning project-level data definitions with enterprise data dictionaries
Resolving conflicts in metric calculation methods across concurrent Six Sigma projects
Standardizing data storage formats for comparative analysis across business units
Implementing metadata management to ensure consistent interpretation of shared datasets
Coordinating with IT to maintain data access permissions and security protocols
Reporting aggregated project outcomes to executive dashboards using consistent KPIs
Managing version control for process maps and data flow diagrams across updates
Conducting lessons-learned reviews on data management challenges across completed projects