Description

This curriculum spans the statistical rigor of multi-workshop quality engineering programs, covering the design and audit of control systems, experiments, and sampling plans as applied in regulated manufacturing and quality assurance environments.

Module 1: Foundations of Statistical Quality Control in Regulated Environments

Selecting between attribute and variable control charts based on measurement system capability and process data type.
Defining rational subgroups in batch manufacturing to ensure within-group homogeneity and between-group variability detection.
Implementing pre-control charts in high-mix, low-volume production where traditional SPC may not be feasible.
Validating normality assumptions before applying parametric control limits; choosing appropriate transformations or non-parametric alternatives.
Aligning control chart frequency with production cycle times to avoid over-sampling or missing critical process shifts.
Documenting statistical rationale for control limit calculations to satisfy ISO 13485 or IATF 16949 audit requirements.

Module 2: Measurement Systems Analysis and Gage R&R Execution

Designing crossed vs. nested Gage R&R studies based on operator-part interaction and destructive testing constraints.
Interpreting %Tolerance, %Study Variation, and %Process metrics to determine whether a measurement system is acceptable, conditionally acceptable, or unacceptable.
Integrating MSA results into calibration schedules and operator training when repeatability exceeds 20% of tolerance.
Handling non-normal measurement error distributions by applying non-parametric confidence intervals or bootstrapping methods.
Coordinating gage R&R with engineering change control when new fixtures or digital measurement tools are introduced.
Establishing acceptance criteria for attribute agreement analysis in visual inspection processes with multiple graders.

Module 3: Process Capability and Performance Index Applications

Differentiating between Cp/Cpk and Pp/Ppk based on process stability and using control charts to validate the assumption.
Calculating capability indices for one-sided specifications, such as flatness or surface roughness, using Cpk(lower) or Cpk(upper).
Adjusting sampling strategy when short-term capability (Cpk) is acceptable but long-term performance (Ppk) deteriorates.
Handling non-normal data using Weibull or log-normal distributions and validating fit with Anderson-Darling tests.
Setting minimum capability thresholds (e.g., Cpk ≥ 1.33) in supplier scorecards and managing exceptions through containment plans.
Linking process capability results to FMEA severity ratings to prioritize improvement efforts in high-risk operations.

Module 4: Design and Analysis of Industrial Experiments (DOE)

Selecting full factorial, fractional factorial, or response surface designs based on resource constraints and interaction effects of interest.
Randomizing run order in production environments to mitigate time-based confounding factors like tool wear or shift changes.
Blocking experiments by lot or machine when uncontrollable noise variables cannot be eliminated.
Interpreting interaction plots to identify non-additive effects between process parameters such as temperature and pressure.
Validating model residuals for homoscedasticity and normality before drawing conclusions from ANOVA results.
Translating statistically significant factors into control plan updates and standard operating procedures.

Module 5: Advanced Control Charting and Real-Time Monitoring

Implementing EWMA or CUSUM charts for early detection of small process shifts in continuous pharmaceutical manufacturing.
Configuring automated SPC software to trigger alerts based on Westgard or Nelson rules without causing alarm fatigue.
Managing multivariate processes using T² charts and interpreting contributions of individual variables to out-of-control signals.
Integrating control charts with SCADA systems to enable real-time monitoring of critical process parameters.
Handling autocorrelated data in high-frequency sensor readings by applying time-series modeling or rational subgrouping.
Defining escalation protocols for out-of-control conditions, including immediate containment and root cause analysis triggers.

Module 6: Statistical Aspects of Nonconformance and Corrective Action

Using Pareto analysis on defect codes to prioritize CAPA efforts while adjusting for sample size differences across lines.
Applying chi-square tests to determine if defect rates differ significantly across shifts, machines, or material lots.
Calculating confidence intervals around defect rates to assess whether observed improvements are statistically significant.
Designing statistically valid sampling plans for rework verification after a nonconformance event.
Linking trended nonconformance data to management review inputs under ISO 9001 clause 9.3.
Using logistic regression to model probability of defect occurrence based on process input variables.

Module 7: Sampling Plans and Acceptance Testing in Quality Systems

Selecting between ANSI/ASQ Z1.4, Z1.9, or custom variables sampling plans based on product risk and inspection cost.
Calculating AQL and LTPD values in alignment with customer requirements and field failure cost models.
Transitioning between normal, tightened, and reduced inspection based on predefined switching rules and supplier performance.
Validating zero-defect sampling plans using OC curves to communicate risk to stakeholders.
Applying double or sequential sampling to reduce average sample size in high-volume incoming inspection.
Documenting statistical justification for reduced or eliminated inspection under continuous process verification programs.

Module 8: Integration of Statistical Methods into Quality Management System Audits

Reviewing control chart archives during internal audits to verify consistent application of SPC rules across production areas.
Assessing adequacy of statistical justification in deviation investigations involving process drift or out-of-specification results.
Evaluating whether MSA and capability studies are repeated at appropriate intervals defined in the quality plan.
Verifying that statistical software used in quality decisions is validated per 21 CFR Part 11 or Annex 11 requirements.
Checking alignment between statistical conclusions in CAPA reports and actual process changes implemented on the floor.
Training auditors to recognize misuse of statistics, such as treating correlation as causation in root cause analysis.