Focused capabilities that turn manufacturing, test, and reliability data into defensible
conclusions — chosen and documented to hold up in audits and submissions.
Design of Experiments (DOE)
Aligned with ASTM & Six Sigma DMAIC practice
Structured screening, optimization, and robustness studies that reveal how factors and interactions affect product and process performance — with the fewest runs needed to answer the question.
What you can expect
Screening, full/fractional factorial, and response-surface designs
Power and run-size justification before you spend bench time
Interaction and curvature analysis with actionable models
Robustness and optimization studies to set operating windows
Process Validation Statistics
Supports FDA Process Validation guidance, ICH Q8–Q11
Statistical strategy and analysis for process design, qualification, and continued verification — sampling plans, acceptance criteria, and evidence that your process is capable and in control.
What you can expect
Stage 1–3 statistical strategy and protocols
Sampling plans and acceptance criteria with statistical basis
PQ/OQ data analysis and capability demonstration
Continued Process Verification (CPV) monitoring plans
Statistical Process Control (SPC)
Aligned with ASTM E2587 and industry SPC practice
Design and implementation of control charting and monitoring systems that distinguish common-cause variation from special causes — so your team reacts to signals, not noise.
What you can expect
Control chart selection, limits, and rule sets
Subgrouping and rational sampling strategy
Monitoring dashboards and out-of-control action plans
CPV and post-market trending programs
Process Capability & Tolerance Analysis
Capability and performance analysis (Cp, Cpk, Pp, Ppk) with the distributional checks behind them, plus statistical and tolerance intervals that quantify how much of your output truly conforms.
What you can expect
Cp / Cpk / Pp / Ppk with normality and stability checks
Non-normal and transformation-based capability methods
Statistical, tolerance, and prediction intervals
Specification and tolerance stack-up support
Measurement Systems Analysis (MSA)
Aligned with AIAG MSA and test method validation practice
Gauge R&R and measurement-system studies that separate product variation from measurement variation — verifying that your inspection and test methods are fit for their purpose.
What you can expect
Gauge R&R (variable and attribute) study design and analysis
Bias, linearity, and stability assessment
Test method validation (TMV) support
Guidance to improve and re-qualify measurement systems
Acceptance Sampling Plans
ANSI/ASQ Z1.4, Z1.9, and zero-acceptance methods
Sampling plans for incoming, in-process, and final inspection built on explicit producer and consumer risk — including zero-acceptance and attribute/variables plans matched to your quality targets.
What you can expect
ANSI/ASQ Z1.4 and Z1.9 plan selection
C = 0 and reduced-inspection strategies
OC-curve analysis with AQL / RQL and risk trade-offs
Sampling rationale for design controls and QMS records
Sample Size & Power
Sample-size and power justification for verification, validation, and reliability studies, so your studies are adequately powered and your conclusions withstand review.
What you can expect
Sample-size and power calculations with documented assumptions
Study design review for V&V and design verification
Reviewer-ready statistical rationale memos
Reliability & Life Data Analysis
Supports ASTM & ICH Q1E stability practice
Life-data and reliability modeling for product durability, shelf-life, and stability — turning time-to-event and stability data into estimates you can defend to reviewers and customers.
What you can expect
Weibull and life-data (parametric) modeling
Reliability demonstration and confidence bounds
Shelf-life and stability estimation for product data
Accelerated testing and degradation analysis support
Statistical Programming & Applications
Version-controlled and documented to support software validation
Statistical programming and application development in R and Python — from reproducible, validated analysis code to interactive Shiny and Streamlit apps your team can run on their own. We also extend Minitab and JMP with custom macros and scripts, automating the analyses you run again and again.
What you can expect
Reproducible analysis code and pipelines in R and Python
Interactive web apps built with Shiny (R) and Streamlit (Python)
Deployment and handoff so your team runs analyses independently
Custom Minitab macros and JMP scripts (JSL) to automate routine work
Statistical SOP Development
Fits your QMS documentation (ISO 13485, 21 CFR 820)
Development of statistical standard operating procedures and work instructions — codifying how your teams choose methods, size studies, set acceptance criteria, and interpret results, so practice stays consistent and inspection-ready across projects and people.
What you can expect
Statistical SOPs and work instructions
Method-selection and sample-size decision guides
Acceptance-criteria and data-handling standards
Templates and worked examples your teams can reuse
Statistical Training
Customized to your team’s roles and tools
Targeted training that helps engineers, quality professionals, and scientists apply statistics correctly in their day-to-day work — from DOE and SPC to capability and sampling — taught with your own processes and data as the examples.
What you can expect
Applied workshops tailored to your team and products
Hands-on sessions in your software (R, JMP, Minitab)
Focused modules: DOE, SPC, capability, MSA, and sampling
Practical reference materials your team keeps
A note on scope
M-D Stats focuses exclusively on the statistical analysis of product and process data.
We do not provide clinical trial design or clinical data analysis. If your question sits
outside our focus, we’ll tell you plainly and, where we can, point you in the right direction.
Start a conversation
Not sure which analysis your problem needs?
Describe the data and the decision behind it. We’ll recommend the right statistical approach — and the evidence your reviewers will look for.
