Lesson 2: Processes, CTQs, cost of poor quality, Pareto, and the basic metrics

Lesson 1 covered what Six Sigma is and who runs it. This lesson covers the working vocabulary every belt uses to describe a process and measure how badly, or how well, it is performing: what a process actually is, what "critical to quality" means, what poor quality costs, how to find the few problems worth chasing first, and the handful of metrics that show up on nearly every Six Sigma project.

Defining a process

A process is any set of activities that takes inputs and turns them into an output for a customer, internal or external. Every process can be described the same way: it has suppliers, inputs, a process, outputs, and customers, an ordering usually shortened to SIPOC, covered fully as a mapping tool in Lesson 5. At this stage, the point is simpler: a process is observable, repeatable, and measurable, which is exactly what makes it improvable. A one-time event is not a process. A recurring activity with identifiable steps, inputs, and outputs is.

Critical to Quality characteristics (CTQs)

A CTQ is a measurable characteristic of a product or process whose performance standards must be met to satisfy the customer. CTQs translate a vague customer requirement, such as "the report needs to be fast," into something a team can actually measure and improve, such as "the report must be delivered within two business days of request, at least 95 percent of the time." Every Six Sigma project needs at least one clearly defined CTQ, because without one there is no way to know objectively whether the project succeeded.

Cost of Poor Quality (COPQ)

Cost of Poor Quality is the total cost an organization incurs because its process, product, or service was not done right the first time. COPQ is usually broken into categories: internal failure costs, caught before the customer sees them, such as rework and scrap; external failure costs, caught after the customer sees them, such as complaints, returns, and warranty claims, which are typically far more expensive than internal failures; appraisal costs, spent finding defects, such as inspection; and prevention costs, spent avoiding defects in the first place, such as training and process design. COPQ is one of the most persuasive tools a Yellow Belt has for building a business case, because it translates a quality problem directly into a dollar figure leadership can act on.

Pareto analysis, the 80:20 rule

Pareto analysis is built on the observation, generalized from economist Vilfredo Pareto's original study of wealth distribution, that a small proportion of causes is typically responsible for a large proportion of effects, commonly stated as roughly 80 percent of the problem coming from roughly 20 percent of the causes. A Pareto chart is a bar chart of defect or cost categories ranked from largest to smallest, usually with a cumulative percentage line overlaid, that makes the vital few causes visually obvious against the trivial many. The tool's purpose is prioritization: a team with limited time should attack the tallest bars first, not spread effort evenly across every category with equal weight.

The basic Six Sigma metrics

A handful of metrics recur constantly across Six Sigma projects, and a Yellow Belt needs to know what each one measures and, at a basic level, how it is calculated:

• DPU, Defects Per Unit. The total number of defects found divided by the total number of units produced. A batch of 100 units with 8 total defects found across them has a DPU of 0.08. Note that DPU counts defects, not defective units; a single unit can have more than one defect.
• DPMO, Defects Per Million Opportunities. DPU scaled to a common, comparable basis: the number of defects divided by the number of units times the number of opportunities for a defect per unit, multiplied by one million. DPMO is what allows a process making one product with five defect opportunities per unit to be compared fairly against a process making a very different product with twenty opportunities per unit.
• FTY, First Time Yield. The percentage of units that pass through a single process step correctly the first time, with no rework, scrap, or correction needed. A step that produces 100 units, of which 92 pass with no rework, has an FTY of 92 percent.
• RTY, Rolled Throughput Yield. The probability that a unit will pass through every step of a multi-step process with zero defects at any step, calculated by multiplying the FTY of each individual step together. A three-step process with FTYs of 95 percent, 90 percent, and 98 percent has an RTY of 0.95 times 0.90 times 0.98, about 83.8 percent, well below any single step's own yield. RTY is why a process can have every individual step looking healthy and still deliver a poor overall result: the multiplication compounds every step's imperfection.
• Cycle Time. The total elapsed time for one unit to move from the start of a process to its completion. (Some Lean texts reserve cycle time for the time between finished units at one step, and use lead time for a customer's wait from request to delivery; this course uses the start-to-completion meaning.) Cycle time is frequently the Y in a Lean Six Sigma project aimed at Lean's core goal of eliminating wasted time, covered in Lesson 4.

These five metrics are not independent trivia. They are the standard language a Yellow Belt uses to describe how a process is performing before a project ever tries to change it, and the Measure phase lessons later in this course build directly on them.