Process Capability Analysis

Process capability analysis is a statistical tool used in Six Sigma to measure how well a process can meet specified requirements. It compares the natural variation of a process with the specification limits to determine if the process is capable of producing output within those limits.

What is Process Capability?

Process capability refers to the ability of a process to produce output that meets customer requirements or specifications. It's measured by comparing the process spread (6σ) with the specification width (USL - LSL).

Key Concepts:

  • Specification Limits: The acceptable range for a product or process characteristic
  • Process Spread: The natural variation in the process (typically ±3σ)
  • Capability Indices: Numerical measures of process capability (Cp, Cpk, Pp, Ppk)

Process Capability Indices

Cp (Process Capability)

Cp measures the potential capability of a process, assuming the process is centered between specification limits:

Cp = (USL - LSL) / (6σ)

  • Cp < 1: Process not capable
  • Cp = 1: Process barely capable
  • Cp > 1: Process capable

Cpk (Process Capability Index)

Cpk measures the actual capability of a process, taking into account how centered the process is:

Cpk = min[(USL - μ) / (3σ), (μ - LSL) / (3σ)]

  • Cpk < 1: Process not capable
  • Cpk = 1: Process meets specifications
  • Cpk > 1.33: Good process capability
  • Cpk > 1.67: Excellent process capability

When to Use Process Capability Analysis

  • During the Control phase of DMAIC
  • When establishing process baselines
  • After process improvements
  • For supplier qualification
  • During new product introduction

Steps for Process Capability Analysis

  1. Collect sufficient data (typically 30+ data points)
  2. Verify process stability using control charts
  3. Check data for normality
  4. Calculate process mean and standard deviation
  5. Determine specification limits
  6. Calculate capability indices
  7. Interpret results and take action

Common Challenges

  • Non-normal data distribution
  • Insufficient data
  • Unstable processes
  • Incorrect specification limits
  • Measurement system variation

Related Tools & Resources