Statistical Process Control Graphs
Statistical process control graphs, also called control charts, plot a factory process metric over time with a center line and control limits so readers can tell routine variation from unusual signals.
What It Is
Statistical process control (SPC) uses control charts to watch a process over time. A control chart is a time-ordered plot of a process metric with a center line for the typical level and control limits that mark the band of routine variation. Points inside the limits usually reflect common-cause variation — the ordinary noise of a stable process. Points that break the limits, or form clear non-random patterns, are special-cause signals: unusual shifts that deserve a closer look.Why It Matters
Long-running agent factory work keeps producing metrics every interval: throughput, failure rate, queue depth, and token spend. Without control limits, every bump can look like an incident and every quiet stretch can look healthy. Control charts help operators spot real drift during a multi-hour run — for example a rising failure rate or a jump in token spend — and tell that signal apart from ordinary process noise before anyone treats the blip as a special cause.Simple Example
Imagine counting successful goal completions in each ten-minute interval while a factory run is active. Most intervals land near the same center line and stay between the upper and lower control limits. One interval then spikes above the upper limit after a bad retry storm. The control chart below makes that special-cause excursion visible against the earlier in-control stretch, so the reader can tell unusual behavior from ordinary jitter.Control chart: goal completions per ten-minute interval
X axis: Interval (ten-minute ticks). Y axis: Completions.
- Completions
- Center line
- Upper control limit
- Lower control limit