Course contentsShow
← Data Science
Lesson 1400 of 2,14535. Statistical Anomaly DetectionFree lesson

Control Limits vs Specification Limits

Distinguish between control limits (voice of process) and specification limits (voice of customer).

Control Limits vs Specification Limits

You'll learn the critical difference between what your process actually does (control limits) and what your customer needs (specification limits).

The Core Distinction

In statistical process control, these two types of limits serve completely different purposes:

Control Limits are the "voice of the process." They tell you what your process is naturally capable of doing right now, based on its inherent variation. Think of control limits as describing your car's actual fuel efficiency—maybe it gets 28-32 mpg based on how it currently runs.

Specification Limits are the "voice of the customer." They define acceptable performance boundaries set by customer requirements, engineering standards, or business rules. Using the car analogy, this would be the minimum fuel efficiency (say, 30 mpg) promised in your warranty or required by regulations.

Why This Matters

A process can be in statistical control (staying within control limits) while simultaneously producing defects (falling outside specification limits). Your car might consistently get 29 mpg (predictable, in control), but still fail to meet the 30 mpg specification.

Conversely, a process might meet specifications but be out of control—varying unpredictably. This is dangerous because even if products are acceptable today, the unstable process could produce defects tomorrow.

The Key Relationship

  • Control limits: Calculated from process data (typically ±3σ from the mean)
  • Specification limits: Set externally by requirements
  • Goal: Make your control limits narrower than your specification limits, ensuring consistent quality

Key Takeaway: Control limits describe what your process does naturally; specification limits define what it should achieve. Process improvement means tightening control limits to reliably stay within specification limits.