LimitLine

Learn · updated 2026-10-28

G Chart: A Control Chart for Rare Events

By Nicholas Williford, Lean Six Sigma Black Belt · About LimitLine

A G chart is a control chart of the number of days (or units) between rare events, with limits from the geometric distribution, used when events are too rare for a c or p chart.

When events are rare, most samples count zero and a p or c chart goes flat and useless. The G chart turns the data around: instead of counting events per period, it plots the gap between one event and the next, so every event is a point and a long gap is good news.

Centerḡ = average gap between eventsUCLḡ + 3√(ḡ(ḡ + 1))LCLusually 0 (the formula goes negative)Basisthe geometric distribution (Benneyan)

Worked Example

Sixteen gaps, in days, between recordable injuries at one site (an illustrative example, not real data). The average gap is 30.75 days, so UCL = 30.75 + 3√(30.75 × 31.75) = 124.5 days.

UCL 124.5LCL 0.0CL 30.8Point 1: 12Point 2: 30Point 3: 5Point 4: 41Point 5: 18Point 6: 9Point 7: 60Point 8: 22Point 9: 3Point 10: 35Point 11: 15Point 12: 48Point 13: 27Point 14: 8Point 15: 19Point 16: 140 (Signal)13579111316
The last gap, 140 days, is above the UCL. On a G chart a point above the UCL is the improvement you were hoping for.

Two cautions before you celebrate. First, set the limits without the suspect point: from the first 15 gaps the mean is 23.5 days and the UCL 95.4, and 140 is still above it, so the signal holds. Second, the gaps follow a skewed (geometric) distribution, so 3σ limits give false alarms on about 2% of points rather than 0.27%. Read one point above the UCL as "the rate has probably dropped", and confirm it with the next few gaps (or use probability limits).

How to Read It

Hospitals use G charts for infections and falls, plants for injuries and line stoppages, and IT teams for outages: anything rare enough that most weeks have none.

Draw a G chart in LimitLineAdvanced control charts include the G chart for days between events (a Pro tool; Free users can try it on the example).

Sources

  1. J. C. Benneyan (2001), Number-Between g-Type Statistical Quality Control Charts for Monitoring Adverse Events, Health Care Management Science 4, 305–318
  2. D. C. Montgomery, Introduction to Statistical Quality Control, Wiley

The LimitLine Newsletter

About one email a month: new guides, tool updates and one practical SPC tip. Unsubscribe in one click.

We'll email you once to confirm before anything else is sent. Privacy