Without a baselineThe established, quantified level of current process performance for a specific metric (the Y) before any improvement efforts are implemented. It is the definitive benchmark against which all future results will be compared, ensuring the project's success and Return on Investment (ROI) are measured ... Click term for full definition, you can never prove improvement. This guide brings all key formulas, the Sigma table and a baseline statement together on a few pages. Watch the short video, then work with the toolkit PDF: it explains the method, shows a worked example and gives you a blank template to fill in.
Tool video
The short tool video will appear here.
Baseline & Sigma Level in short
- Check your data type. Counted (defectsAn error, flaw, or non-conformance in a product or service that fails to meet a specific Critical-to-Quality (CTQ) requirement or Specification Limit. A single unit or service can contain multiple defects. In Six Sigma, the goal is to define, measure, and eliminate these occurrences. Click term for full definition, errors) or measured (time, weight)?
- Discrete: calculate DPMODefects Per Million Opportunities: the number of defects in a process per one million chances for a defect to occur. It is the statistical measure behind the Six Sigma name (3.4 DPMO). Click term for full definition. Count defects, units and opportunities per unit (OPU).
- Look up the Sigma LevelThe definitive measure of process capability, expressing how many standard deviations fit between the process performance and the customer's specification limit. Also known as: Z-Score. Click term for full definition. Use the DPMO-to-Sigma table in this guide.
- Continuous: calculate Z. Z = (mean − nearest spec limit) ÷ σ, then subtract the 1.5 shift.
- Write the baseline statement. What, how much, how much variation, over which period.
Key points
- Baseline = average + variation + period.
- Discrete: DPMO, then look up Sigma.
- Continuous: Z-score against the spec limit.
- Your benchmark for proving success.