Dictionary of Lean Six Sigma Tools/Concepts
There are 11 entries in this dictionary.| Term | Definition |
|---|---|
| Data Collection Plan (DCP) | A detailed plan describing exactly what data elements are necessary and who will acquire the data in order to properly measure the potential root causes that were identified. The purpose is to define what data is to be collected and the precise method for collecting it, to unify the team around the purpose for collecting the data and how it aligns to the causes, and to keep the team accountable for ensuring the data is collected in the prescribed manner. |
| Defects/Defectives | An undesirable effect that occurs in a process outside of what the customer wants (a.k.a. the voice of the customer or VOC) as defined in a process as a lower/upper specification limit (LSL/USL). Something defective typically reflects the severity of how defective a unit is (i.e., by the number of defects it contains). |
| Deviation | The distance a data point is from the mean (a.k.a. average) or (X - μ). |
| DFSS (Design for Six Sigma) | "Design for Six Sigma" is a Six Sigma methodology that is generally used for creating a new output (like a product) that is intended to follow the Six Sigma standards of preventing defects and having high quality or accuracy. |
| Discrete Data | Also known as attribute or categorical data, it's a type of data that can be measured by counts or classifications that are limited in scale and divisibility. General examples include yes/no, pass/fail, colors, locations, proportion values (based on limited counts), etc. Business examples include people (names or roles), product types, stores/offices, defects, % performance, etc. This can be contrasted with Continuous Data. |
| Distributions | A statistical representation plotting all continuous (a.k.a. variable or numerical) values or data points of a dataset along a continuous scale. The plotting of these observations create a shape often referred to as a "bell curve" because of the appearance like a bell. The statistical measurement of the shape of the distribution is defined as normality, i.e., a normal distribution or non-normal distribution. |
| DMADV | A Six Sigma methodology that consists of five phases that make up its namesake: Define, Measure, Analyze, Design, and Verify. It is generally used for designing a new efficient process that is intended to follow the Six Sigma standards of preventing non-value-added steps and planning for optimal flow and speed. |
| DMAIC | The 5 phases one of the most common methodologies of Six Sigma. The phases are Define, Measure, Analyze, Improve and Control. It is essentially like the scientific method adapted to businesses for resolving problems using data and analysis. |
| DMAIC Roadmaps | A guide that lists questions addressing the critical requirements for each of the 5 phases in the DMAIC methodology. There are 3 levels of the roadmap that cover 1) a high level overview, 2) a mid level detail, and 3) a low, very detailed content. Levels 2 & 3 of the roadmap include a list of tools and resources that may be used to help answer each respective question, and a list of tools and resources that may normally serve as the output from each respective phase. |
| DOE (Design of Experiments) | A detailed method for designing tests or experiments for measuring variation in a process. It's generally included as a part of statistical process control (SPC). |
| DPMO | It stands for Defects per Million Opportunities and is a count of the number of defects expected to occur for every one million opportunities run through the process. It's essentially like a percent defective or p(d) that's carried out to the 4th decimal place. The equation to calculate it is (Total Defects / (Total Units x Opportunities per Unit)) x 1,000,000. |
Search the Dictionary of Lean Six Sigma Tools/Concepts
