Quality
Pre-Control: When Your Shop Floor Stops Calculating Sigma and Starts
Catching Defects With Three Lines and a Ruler — and the Simplest
Statistical Method You’ve Been Ignoring Outperforms Your Most Complex
Control Chart
You spent six months training your operators on X-bar and R charts.
You bought the software. You printed the forms. You explained upper
control limits, lower control limits, runs, trends, and out-of-control
signals. You felt proud.
Then you walked the floor on a Tuesday afternoon and watched an
operator fill in twelve data points from memory because he was behind on
production. Another operator had drawn a perfect bell curve on his chart
— every point equidistant, every trend absent — because his supervisor
told him the customer audits the charts and “they need to look
clean.”
Your control charts were works of art. Your process was a
disaster.
This is not a story about bad operators. This is a story about a tool
that was too complex for the people who needed to use it — and about a
simpler method that has been sitting in the quality engineer’s toolbox
since 1957, waiting for the moment when you’d finally get tired of
charting data nobody reads.
That method is Pre-Control. And it might be the most powerful quality
tool you’ve never seriously considered.
What Is Pre-Control?
Pre-Control is a statistical process control technique developed by
Frank Satterthwaite at Rath & Strong Consulting in the 1950s. It was
designed to answer a deceptively simple question: Is my process
capable of producing conforming parts right now, with the next
piece?
Not “what was my Cpk last month?” Not “how many sigma am I from the
mean?” But a real-time, piece-by-piece, go/no-go decision that any
operator can make in seconds.
It uses three zones, defined by two lines drawn on the tolerance
range:
- The Green Zone — the central half of the tolerance
range. If your part lands here, keep running. Everything is fine. - The Yellow Zone — the outer quarters of the
tolerance range, between the green zone boundary and the specification
limit. A part here is still conforming, but it’s a warning. You’re
drifting. - The Red Zone — anything outside the specification
limits. Stop. Investigate. Adjust.
That’s it. Three colors. No calculations. No standard deviations. No
looking up D4 or A2 constants in a table you can never find when you
need it.
The genius of Pre-Control is not that it’s simple. The genius is that
it’s simple AND statistically rigorous.
The Rules: How
Pre-Control Actually Works
Pre-Control operates on a set of decision rules that take about five
minutes to teach and about five seconds to apply.
Setup Qualification
(Starting a Run)
Before you begin production, you measure five consecutive parts:
- All five in the Green Zone? You’re qualified. Start
production. - One in the Yellow? Keep measuring. You need five
consecutive in Green. - Any in the Red? Stop. Your process is not capable
of running. Investigate and fix before you start.
This setup rule is remarkably powerful. If your process is centered
and capable (Cpk ≥ 1.33), the probability of getting five consecutive
parts in the Green Zone is high. If your process is marginal or
off-center, it will fail this test immediately — before you produce a
single nonconforming part for your customer.
During Production
(Monitoring)
Once running, you sample two consecutive parts at regular
intervals:
- Both Green: Continue running.
- First Green, Second Yellow: Continue, but you’re on
notice. Drift detected. - First Yellow, Second Yellow (same side): Stop.
Adjust the process to re-center it. - First Yellow, Second Yellow (opposite sides): Stop.
Your process variability has increased. You have a dispersion problem,
not a centering problem. This requires a different kind of
investigation. - Either part Red: Stop immediately.
Investigate.
After any adjustment, you re-qualify with five consecutive parts in
the Green Zone.
These rules create a self-correcting system. The operator doesn’t
need to understand statistics. They need to understand colors: green
means go, yellow means caution, red means stop.
Why
Pre-Control Works: The Statistical Foundation
I know what you’re thinking. “This sounds too simple to be
statistically valid.”
It’s not. And the mathematics behind it are elegant.
Pre-Control is built on the assumption that your process is
approximately normally distributed — the same assumption underlying
traditional Shewhart control charts. The zone boundaries are positioned
so that:
- For a capable process (Cpk = 1.33), approximately 88% of
parts fall in the Green Zone, 12% in the
Yellow, and virtually none in the Red. - For a marginal process (Cpk = 1.0), the Yellow zone catches drift
before the Red zone is reached. - For an incapable process, the setup qualification fails immediately,
preventing production from starting.
The two-part sampling rule is the key statistical mechanism:
- The probability of two consecutive Yellow parts on the same
side when the process is centered and capable is roughly
1 in 72 — a rare event that signals a genuine mean
shift. - The probability of two consecutive Yellow parts on opposite
sides signals increased spread — a different problem requiring
a different response. - The probability of a Red part from a capable process is so low that
a single occurrence warrants immediate investigation.
These aren’t arbitrary rules. They are statistically designed
tripwires calibrated to balance two risks: the risk of missing a
real problem (Type II error) and the risk of stopping
production unnecessarily (Type I error).
The result is a system that catches about 85-90% of process
shifts — comparable to an X-bar chart with a sample size of 4-5
— while being usable by someone who has never heard of Walter
Shewhart.
The Setup
Phase: Your Most Undervalued Quality Gate
Most quality professionals focus on monitoring. Pre-Control focuses
first on qualification — and this is where it delivers
disproportionate value.
Think about your current process. A new job starts. The operator sets
up the machine, runs a few parts, measures them, and if they’re within
spec, production begins. How many parts do they measure? Three? Five? Do
they check centering? Do they verify capability?
In most shops, the answer is: they check if the parts pass, and if
they do, they run. The process could be centered near the upper
specification limit, one tiny drift away from producing scrap, and
nobody would know until the scrap appears.
Pre-Control’s five-piece setup rule changes this completely. Five
consecutive parts in the Green Zone means your process is not just
within spec — it’s centered and capable. You’ve proven,
before producing a single production part, that your process has enough
room on both sides to absorb normal variation.
This single rule has prevented more startup defects than any control
chart I’ve ever implemented. I’ve seen it happen dozens of times: an
operator sets up a new job, measures five parts, and can’t get all five
in the Green Zone. The process keeps bouncing into the Yellow. Without
Pre-Control, those parts would have passed inspection. The operator
would have started running. And within an hour, the drift would have
pushed them into the Red.
Instead, the setup rule caught the problem before it started. Every
time.
Pre-Control
vs. Shewhart Charts: An Honest Comparison
Let me be clear about something: Pre-Control is not a replacement for
Shewhart control charts in every situation. Each tool has its sweet
spot.
When Pre-Control Is Better
Short production runs. If you’re running 50 parts,
you’ll never accumulate enough data for a meaningful control chart.
Pre-Control gives you statistical protection from part one.
High-mix, low-volume environments. Job shops,
contract manufacturers, custom fabrication — these environments change
jobs constantly. By the time you’ve established control limits, the job
is over. Pre-Control’s setup qualification gives you immediate
protection.
Operator-driven processes. Machining, assembly,
welding — processes where the operator is the primary source of
variation and the one making real-time adjustments. Pre-Control puts the
decision in the operator’s hands without requiring statistical
training.
Processes with frequent setups. If you change over
multiple times per shift, you need a method that qualifies each setup
quickly and reliably. Five parts in the Green Zone takes about two
minutes.
When Shewhart Charts Are
Better
Long, stable production runs. If you’re running the
same part for months, Shewhart charts give you richer information about
trends, cycles, and subtle shifts.
Regulated industries with specific requirements.
Some customer and regulatory requirements explicitly mandate control
charts. Pre-Control may supplement but not replace them.
Root cause analysis. When you need to understand the
behavior of a process over time — identifying patterns, correlating with
inputs, understanding sources of variation — Shewhart charts with their
detailed historical record are superior.
Process capability studies. If you need to calculate
Cpk, Ppk, or conduct a formal process study, you need the data that
Shewhart charts provide.
The honest quality engineer doesn’t choose one over the other. They
choose the right tool for the situation. And in more situations than
most people realize, Pre-Control is the right tool.
Implementation: A Practical
Guide
Implementing Pre-Control is straightforward, but like any quality
tool, the details matter.
Step 1: Identify Your
Characteristic
Not every dimension needs Pre-Control. Apply it to your critical
characteristics — the dimensions, properties, or features that most
directly affect product function and customer satisfaction. These are
the same characteristics you’d identify in your control plan.
Step 2: Define the Zones
Take your specification limits (USL and LSL). Calculate the midpoint.
The Green Zone extends from the midpoint outward to ¼ of the total
tolerance on each side. The Yellow Zone extends from the Green Zone
boundary to the specification limit. The Red Zone is everything
beyond.
Example: Your specification is 10.000 ± 0.010 mm. – USL = 10.010, LSL
= 9.990 – Midpoint = 10.000 – Total tolerance = 0.020 – Green Zone:
10.000 ± 0.005 = 9.995 to 10.005 – Yellow Zones: 9.990 to 9.995 (lower)
and 10.005 to 10.010 (upper) – Red Zones: below 9.990 and above
10.010
Draw these on a simple chart next to the workstation. Color them.
Make them visible from three meters away.
Step 3: Train the Operators
This should take 15 minutes, not 15 hours. Show them the three zones.
Explain the setup rule (five green to start) and the monitoring rule
(two-part sample, color-coded decisions). Role-play a few scenarios.
Post a one-page reference guide at the workstation.
The most important training point: stopping is not a
punishment. If the culture punishes operators for stopping
production, Pre-Control will be ignored just like every other quality
tool you’ve tried to implement on that floor.
Step 4: Set the Sampling
Frequency
Pre-Control doesn’t prescribe a specific sampling interval. Use your
judgment based on process stability and production volume. Common
starting points:
- Every 25-50 parts for stable, high-volume processes
- Every 5-10 parts for less stable or critical processes
- After every tool change, material change, or shift change
Step 5: Respond to the
Signals
This is where most implementations fail. Pre-Control will tell you
when to stop. But if nobody responds to the stop signal — if the
operator adjusts and re-qualifies but the root cause is never
investigated — you’re just playing a game of whack-a-mole.
Every Yellow-Yellow or Red stop should be logged. Every adjustment
should be recorded. And periodically, you should review these events to
identify recurring problems that need permanent fixes, not temporary
adjustments.
The Cultural
Advantage Nobody Talks About
Here’s something the textbooks won’t tell you: Pre-Control changes
the conversation on your shop floor.
With traditional SPC, the operator’s job is to record data. The chart
belongs to the quality department. The operator fills in points, and
someone in an office interprets them. The operator is a data entry
clerk.
With Pre-Control, the operator is the decision-maker. They measure a
part. They see a color. They make a decision — run, adjust, or stop — in
real time, on the spot, without calling quality engineering. The quality
tool belongs to the operator, not to the quality department.
This is a profound shift. It transforms quality from something done
to the operator into something done by the operator. It moves quality
from a staff function to a line function. And it communicates something
powerful to every person on your floor: we trust you to make
quality decisions.
I’ve implemented Pre-Control in facilities where SPC had been
“implemented” three times and abandoned three times. Each time, the
reason was the same: the operators didn’t understand it, didn’t own it,
and didn’t use it. Pre-Control succeeded because it was designed for
them — not for the quality engineer’s spreadsheet.
Common Objections (And
Why They’re Wrong)
“Pre-Control isn’t as sensitive as Shewhart
charts.”
True. Shewhart charts with appropriate sample sizes detect smaller
shifts. But sensitivity without action is useless. A control chart that
nobody reads detects nothing. Pre-Control detects shifts that matter,
and it does it in a way that triggers actual human response.
“It’s too old-fashioned. We should be using AI and machine
learning.”
Pre-Control was developed in 1957. The Pythagorean theorem was
developed around 500 BC. Age is not a disqualification. The question is:
does it work? Pre-Control works. And unlike your AI predictive
maintenance system that requires three data scientists and a cloud
subscription, Pre-Control works with a ruler and three colored
markers.
“Our customer requires control charts.”
Then use control charts for your customer. Use Pre-Control for
yourself. They’re not mutually exclusive. Some of the best
implementations I’ve seen run both: Shewhart charts for the customer
documentation package, Pre-Control for real-time shop floor
decisions.
“Our process is too complex for such a simple
method.”
Complexity of the process is not the same as complexity of the
monitoring method. A complex process still produces individual parts
with individual measurements. Those measurements are either centered and
capable, or they’re not. Pre-Control answers that question regardless of
how many variables produced the part.
The Bottom Line
Pre-Control is not the answer to every quality question. But it is
the answer to one of the most common quality problems on the planet:
a sophisticated statistical tool that nobody on the floor uses
correctly.
It sacrifices some analytical depth for enormous gains in usability,
speed, and operator engagement. In environments with short runs,
frequent setups, and operator-driven processes, those gains are not just
nice to have — they’re the difference between a quality system that
lives on paper and one that lives on the floor.
The next time you catch an operator filling in control chart data
from memory, ask yourself: is the problem the operator, or is the
problem the tool? Then draw three colored zones on a piece of paper and
see what happens.
You might be surprised how much quality three lines and a ruler can
deliver.
Peter Stasko is a Quality Architect with over 25
years of experience in automotive and manufacturing quality management.
He has implemented quality systems across three continents, led hundreds
of audits, and believes that the best quality tool is the one people
actually use. His work focuses on bridging the gap between statistical
theory and shop floor reality — because a method that sits in a textbook
helps no one.
