Tag: green-belt

Control limits on a control chart are calculated from the actual process data itself, typically set at ±3 standard deviations from the process mean. They reflect the voice of the process - the natural variation inherent in the process. Specification limits come from customers, not from GB or BB roles.

COPIS (Customer-Output-Process-Input-Supplier) is a process mapping tool with exactly five components that correspond to its name. The five components are Customer, Output, Process, Input, and Supplier. 'Operation' is not one of these components, making C the correct answer for what COPIS does not contain.

Correlation and causation are fundamentally different concepts. Correlation indicates that two variables change together, but does not prove that one variable causes the other to change. A correlation could be coincidental, or both variables might be caused by a third hidden variable (confounding factor). This is why 'Correlation doesn't mean causation' is a TRUE statement.

A Criteria Based Matrix (also called a Decision Matrix or Pugh Matrix) is a decision-making tool used to evaluate multiple solutions against weighted criteria. It helps teams systematically compare alternatives and select the best solution based on objective criteria rather than subjective judgment. The matrix scores each solution against predetermined criteria to identify the optimal choice.

The CTQ/Project Y matrix helps identify the best measurable metrics for Critical to Quality (CTQ) characteristics. While CTQs themselves are customer requirements, this matrix specifically helps determine how to measure each CTQ effectively. It bridges the gap between vague customer needs and specific, measurable metrics (Project Y) that can be tracked and improved.

Data consistency comes from stable measurement systems and consistent procedures, not from using the same inspectors repeatedly. Different trained inspectors following the same documented procedures can produce equally consistent data. Relying on the same inspectors is not necessary and could even introduce bias if that person has consistent measurement habits. Therefore, 'Same inspectors for data collection' is the correct exception.

Attribute data represents categorical or binary outcomes where each observation falls into distinct categories. YES/NO and pass/fail are classic examples of attribute (also called discrete or categorical) data because they have only two possible states. This differs from continuous data which can take any value within a range. Therefore, 'Attribute data' is the correct classification.

A defect is defined as a measurement that falls outside the specification limits set by the customer or design requirements. The Lower Specification Limit (LSL) and Upper Specification Limit (USL) define the acceptable range. Any value below LSL or above USL is a defect. Control limits (LCL, UCL) are different statistical boundaries and do not define defects. Therefore, defects cannot fall between LSL and USL.

The Define phase in Six Sigma DMAIC focuses on clearly defining the problem and project scope. Its three key deliverables are: (1) Project Charter - documenting the problem, goal, scope, and team; (2) CTQs (Critical to Quality) - translating customer needs into measurable requirements; (3) COPIS diagram - mapping the process at a high level from Customer through Supplier. These form the foundation for the remaining phases.

DPU (Defects Per Unit) measures the average number of defects per product or unit. The formula is: DPU = Total Defects ÷ Total Units (or products). This differs from DPO (Defects Per Opportunity) which divides by opportunities, and DPMO which scales to million opportunities. DPU is a straightforward metric that tells you how many defects exist per product on average.