Statistical Process Control (SPC) for Manufacturing

Statistical Process Control (SPC) uses statistical methods to monitor and control production processes. By detecting variation early, SPC prevents defects, reduces scrap, and ensures consistent product quality.

Two Types of Variation

Key SPC Charts

ChartData TypeSample SizeApplication
X-bar & RVariable (continuous)2-10Process mean and range monitoring
X-bar & SVariable (continuous)>10Process mean and std dev monitoring
Individual & MRVariable (continuous)1Slow processes, batch operations
p chartAttribute (defect rate)50-200Fraction non-conforming
np chartAttribute (defect count)50-200Number non-conforming, constant sample
c chartAttribute (defects per unit)Unit areaCount of defects on a surface

Control Limits vs. Specification Limits

Control limits (UCL/LCL) are calculated from process data ±3σ. They define expected process variation. Specification limits (USL/LSL) are engineering requirements for part function. A process can be in control but not capable (producing outside spec). Both metrics matter.

Process Capability Indices

Implementing SPC

  1. Identify critical-to-quality (CTQ) characteristics on the product
  2. Establish measurement system with GR&R analysis (GR&R < 20% of tolerance)
  3. Collect 25+ subgroups to establish baseline control limits
  4. Calculate trial control limits and check for out-of-control points
  5. Remove special causes, recalculate limits
  6. Deploy chart at workstation with clear reaction plan for out-of-control conditions
  7. Review and update limits after 100+ subgroups or after major process changes

Ready to Start Your Project?

Get an instant quote from our engineering team.

Get a Quote →