Quality departments invest heavily in statistical process control (SPC) software and operator training. They deploy X-bar and R charts, define upper and lower control limits, and audit the resulting documentation for IATF 16949 or AS9100 compliance. On paper, the quality system appears robust. On the shop floor, the reality is frequently different.
During plant audits, I regularly find operators filling in historical data points from memory to maintain takt time. Supervisors occasionally draw perfectly distributed bell curves on charts because customer auditors expect clean records. The control charts look immaculate, but the manufacturing process is uncontrolled. The tool has become a documentation exercise rather than a process management instrument.
This failure rarely stems from incompetent operators. It stems from deploying a monitoring tool that requires too much statistical abstraction for the production environment. Pre-control, a statistical method developed by Frank Satterthwaite at Rath & Strong in the 1950s, offers a rigorous alternative. It replaces complex calculations with a piece-by-piece, go/no-go decision framework that any operator can execute in seconds.
The Pre-Control Zone Architecture
Pre-control divides the specification tolerance into three distinct zones, defined by two lines drawn at the midpoint of the acceptable range. It requires no calculation of standard deviation, no lookup of D4 or A2 constants, and no plotting of subgroup averages. The operator simply measures the part and notes which zone the measurement falls into.
The central half of the total tolerance range is the Green Zone. If a part measures within this space, the process is running effectively. The outer quarters of the tolerance range, situated between the Green Zone boundary and the specification limit, constitute the Yellow Zone. A part landing here is conforming, but the measurement acts as a warning that the process mean is drifting. Anything outside the specification limits is the Red Zone. Production must stop immediately.
The statistical validity of pre-control rests on its zone dimensions. Because the Green Zone covers the middle fifty percent of the tolerance, a capable process will naturally produce the vast majority of its parts within these boundaries. The method leverages this statistical distribution without forcing the operator to calculate it.
Pre-Control Probability for a Capable Process
Setup Qualification as a Primary Quality Gate
Most quality professionals focus their effort on in-process monitoring. Pre-control shifts that focus to setup qualification, which delivers disproportionate value in defect prevention. Before starting a production run, the operator must measure five consecutive parts. All five must measure within the Green Zone to authorize production.
If a single part falls into the Yellow Zone, the operator must keep measuring until they achieve five consecutive parts in the Green Zone. If any part falls into the Red Zone, the process is deemed incapable. The operator must investigate, adjust the setup, and re-qualify before producing a single saleable part.
This setup rule acts as a brutal but effective capability filter. If the process is centred and genuinely capable (Cpk ≥ 1.33), passing the five-piece test is statistically straightforward. If the process mean is skewed toward the upper or lower specification limit, the process will fail the test immediately, preventing startup scrap.
In my experience implementing greenfield quality systems, this single rule prevents more startup defects than any Shewhart chart. Operators routinely discover that a nominal setup produces parts hovering near the tolerance limit. Without pre-control, those parts pass initial inspection, and the process drifts into scrap within the first hour of the run.
In-Process Decision Rules
Once production begins, pre-control governs the line through a two-part sampling rule. At regular intervals, the operator measures two consecutive parts. If both parts land in the Green Zone, production continues. If the first part is Green and the second is Yellow, the operator continues running but stays on notice that drift is occurring.
If both parts land in the Yellow Zone on the same side of the target, the operator must stop and adjust the process to re-centre the mean. If both parts land in the Yellow Zone on opposite sides of the target, the operator has a dispersion problem. Increased process variability requires a different investigation than a simple mean shift.
The mathematical mechanism here is the multiplication of independent probabilities. The likelihood of two consecutive parts landing in the same Yellow Zone when the process is properly centred is roughly one in seventy-two. When this rare event occurs, it triggers an immediate human response. If either part measures Red, production stops immediately for a formal 8D root cause analysis.
Sensitivity without action is useless. A control chart that nobody reads detects nothing.
Pre-Control Versus Shewhart Control Charts
Pre-control is not a universal replacement for Shewhart control charts. Shewhart charts remain the superior tool for long, stable production runs where engineers need to identify subtle trends, correlate process inputs, or calculate long-term Cpk and Ppk values for PPAP submissions. They are also mandatory for specific FDA or EASA regulatory compliance documentation.
However, pre-control excels in short production runs and high-mix, low-volume environments. If a job shop runs fifty parts before changing over, operators will never accumulate enough subgroups to establish valid Shewhart control limits. Pre-control provides statistical protection from the very first piece.
Pre-control also dominates in operator-driven processes like machining, assembly, and welding. It places the go/no-go decision directly in the hands of the operator adjusting the machine in real-time, without requiring them to consult a quality engineer.
Tool Selection by Production Environment
When to use Shewhart Charts
- Long, continuous production runs of the same part
- Regulatory or customer mandates for formal SPC documentation
- Deep root cause analysis requiring historical data trends
- Formal long-term process capability studies (Cpk, Ppk)
When to use Pre-Control
- Short production runs and frequent machine changeovers
- High-mix, low-volume job shop manufacturing environments
- Operator-driven processes requiring real-time line adjustment
- Rapid setup qualification before releasing a job to production
Implementation Sequence
Implementing pre-control requires deliberate sequencing. The quality team must first identify the critical characteristics on the control plan. Pre-control is not for every dimension; it belongs on the features that most directly dictate product function, safety, or customer satisfaction.
Next, calculate the zone boundaries using the engineering specifications. For a dimension specified at 10.000 mm ± 0.010 mm, the total tolerance is 0.020 mm. The Green Zone spans the central 0.010 mm (9.995 mm to 10.005 mm). The Yellow Zones occupy the remaining quarters (9.990 mm to 9.995 mm, and 10.005 mm to 10.010 mm). Draw these zones on a physical chart at the workstation, using clear colour coding.
Operator training should take minutes, not hours. Explain the setup rule (five green to start) and the monitoring rule (two-part sample, colour-coded decisions). The most critical training point is cultural: stopping the line for a Yellow or Red result must never be treated as a punishable offense.
Pre-Control Implementation Steps
- 01Identify CharacteristicsSelect critical dimensions from the PFMEA and control plan.
- 02Calculate ZonesDivide the total tolerance into green, yellow, and red boundaries.
- 03Deploy VisuallyCreate colour-coded charts visible from the operator station.
- 04Set FrequencyDetermine sampling intervals based on process stability.
- 05Log ResponsesRecord every stop signal to identify recurring failure modes.
Response Discipline and Cultural Shift
The failure point of any SPC system is response discipline. Pre-control will effectively signal when a process requires adjustment, but if the operator simply tweaks the offset and re-qualifies without investigation, the system degrades into a game of temporary fixes. Every Yellow-Yellow or Red event must be logged for quality engineering review.
When implemented correctly, pre-control fundamentally changes the manufacturing culture. Traditional SPC often relegates operators to data entry clerks for the quality department. The operator records the data, and someone in an office interprets the trends. The operator holds no ownership of the process stability.
Pre-control gives the operator immediate ownership. They measure, identify the colour, and execute the decision. Quality shifts from an administrative overhead applied to the operator into a proactive function executed by the operator. The system communicates a clear operational reality: the people running the machines are trusted to guarantee the quality of the output.