Lean Six Sigma Basics for Quality Professionals

Lean Six Sigma combines two improvement approaches: Lean removes waste so work flows faster, and Six Sigma reduces variation so results are more consistent. Together they give quality teams a structured way to fix recurring problems using data, usually through the DMAIC cycle: Define, Measure, Analyse, Improve and Control.

For quality professionals in laboratories, food plants and pharma, this matters because many audit findings and customer complaints come from the same few process problems repeating. Corrective actions fix single events; Lean Six Sigma is built to fix the process that keeps producing them.

Lean: removing waste

Lean grew out of the Toyota Production System. Its starting point is value, defined from the customer's view, and everything else is waste. The classic Lean principles are to define value, map the value stream, create flow, establish pull and pursue perfection.

The 8 wastes (TIMWOODS)

Value stream mapping, flow and 5S

A value stream map draws every step from request to result, with times and queues, so you can see where waiting and rework happen. Often the actual processing time is a small fraction of the total turnaround time.

5S organises the workplace: Sort (remove what is not needed), Set in order (a place for everything), Shine (clean and inspect), Standardise (agreed ways of working) and Sustain (keep it going through audits and habits). In a laboratory, 5S supports GLP and biosafety as well as efficiency, because a tidy bench is easier to clean and harder to make errors on.

Six Sigma: reducing variation

Six Sigma focuses on variation, because a process that varies widely will sometimes produce defects even when its average is acceptable. It measures performance in defects per million opportunities (DPMO) and uses statistics to find the causes of variation.

A process operating at six sigma produces about 3.4 defects per million opportunities. That figure uses the convention of a 1.5 sigma long-term shift, which assumes process averages drift over time. Common reference points with the same convention are about 66,800 DPMO at three sigma, 6,210 at four sigma and 233 at five sigma.

Process capability compares the spread of a process with its specification limits. Cp compares the specification width with six standard deviations; Cpk also accounts for how far the process mean is off-centre. A Cpk of 1.33 or more is a common target in manufacturing, but targets vary by industry and customer.

In medical laboratories, the sigma metric applies the same thinking to analytical quality, calculated from allowable total error, bias and imprecision, and is used to design QC rules. See our post on Westgard rules for how this links to daily QC.

DMAIC: the core method

Many of these tools are already familiar from quality work. The difference is the discipline: no solution before the baseline is measured, and no root cause accepted without data. Our guides to 5 Whys vs fishbone and FMEA in laboratories cover two of the most used tools.

Belts and roles

• White and Yellow Belt: awareness of concepts; Yellow Belts often support projects as team members
• Green Belt: leads smaller projects alongside a normal job
• Black Belt: leads complex projects, coaches Green Belts and applies advanced statistics
• Master Black Belt: develops the programme, trains and mentors Black Belts
• Champion or sponsor: a manager who owns the business problem and removes barriers

There is no single global authority for belt certification. Requirements, exams and project expectations differ between certifying bodies and employers, so check what a specific certificate involves before enrolling.

When to use Lean, Six Sigma or both

Use Lean when the problem is delay, wasted effort or poor flow, such as long turnaround times or excess handling. Use Six Sigma when the problem is variation or defects, such as inconsistent fill weights or recurring transcription errors. Many real problems involve both, which is why they are combined. For a small, obvious fix, a simple PDCA cycle or a standard CAPA may be enough.

Worked example: reducing sample registration errors in a lab

A hospital laboratory in Abu Dhabi registers around 4,000 test requests a month. Registration errors, such as a wrong test code or missing clinical details, cause repeat collections and delayed reports, and the quality manager sponsors a Green Belt project.

Define. The goal is to cut registration errors by at least half within six months. Scope runs from the request arriving at reception to the sample being labelled in the LIS.

Measure. The team defines five opportunities for error per request: patient identifiers, requesting clinician, test codes, sample type and collection time. That gives 20,000 opportunities a month. A four-week audit finds 96 errors, a baseline of 4,800 DPMO, or roughly 4.1 sigma using the usual convention.

Analyse. A Pareto chart shows most errors are wrong test codes and missing collection times. A fishbone session and observation point to handwritten paper requests, similar-looking codes in the LIS and staff being interrupted at a shared counter.

Improve. The lab works with IT to promote electronic ordering, adds order sets for common panels, makes collection time a mandatory field and creates a quiet registration zone. A two-week pilot is run before full rollout.

Control. A monthly error count is plotted on a control chart, the registration SOP is updated and staff are trained. After three months, errors fall to 18 a month, about 900 DPMO or roughly 4.6 sigma, and the team hands the control plan to the section supervisor.

Common mistakes

• Jumping to solutions before measuring the baseline
• Choosing projects too large to finish in a few months
• Collecting data without checking the measurement system is reliable
• Treating the sigma level as the goal rather than the customer outcome
• Using Lean only as a tidy-up exercise
• Skipping the Control phase, so gains fade

Key takeaway: Lean removes waste and Six Sigma reduces variation. For most quality teams, the real value is the discipline of DMAIC: define the problem clearly, measure before acting, prove root causes with data and build controls so improvements last.

Frequently Asked Questions

What is the difference between Lean and Six Sigma?

Lean focuses on removing waste and improving flow, so processes are faster and use fewer resources. Six Sigma focuses on reducing variation and defects using data and statistics, usually through DMAIC. Lean Six Sigma combines both, because many problems involve delay and inconsistency together. The choice of tools depends on which problem is dominant.

What does six sigma mean in numbers?

A process at six sigma produces about 3.4 defects per million opportunities. This figure assumes a 1.5 sigma long-term shift in the process mean, which is a widely used convention. Without that shift, the theoretical defect rate at six sigma would be far lower. Most real processes, including laboratory and food processes, operate well below six sigma.

What are the 8 wastes in Lean?

The 8 wastes are often remembered as TIMWOODS: transport, inventory, motion, waiting, overproduction, overprocessing, defects and skills, meaning unused talent. The original Toyota model listed seven wastes, and unused skills were added later. Each one consumes time or resources without adding value for the customer.

Is Lean Six Sigma useful for laboratory and food safety roles?

Yes. Laboratories use it to cut turnaround times, pre-analytical errors and rework. Food plants use it to reduce waste, giveaway, rework and recurring nonconformities. It complements ISO 9001, ISO 15189, ISO/IEC 17025 and ISO 22000 systems rather than replacing them, and gives quality staff a structured way to deliver measurable improvement.

Next steps

To connect improvement methods with a working quality system, explore our ISO 9001 Essentials: Quality Management Systems Practical Implementation course, and for problem-solving depth, our Root Cause Analysis & CAPA course. You can also speak to a counsellor.

For related reading, see quality KPIs that actually drive improvement, 8D problem-solving method and turnaround time in clinical labs.