The Tampering Trap: When Fixing Makes It Worse
Interactive simulation showing why over-adjusting a stable manufacturing process increases variation and costs money. Based on Deming's funnel experiment.
Can a careful operator make quality WORSE by trying too hard? Run the simulation to find out.
The Scenario
You run a CNC lathe turning steel shafts. The target outside diameter is 25.000 mm with a tolerance of +/- 0.150 mm. Your process is stable with a natural variation (sigma) of 0.050 mm. Two operators disagree on how to run the machine. Who produces better parts?
AHands Off
Trust the process. Only act when the control chart shows a special cause (point outside control limits).
BOver-Adjuster
Adjust the machine whenever a part measures more than 1 sigma (0.050 mm) from the target. "Better safe than sorry!"
Run the Simulation
---
---
---
---
---
---
Key Takeaways
Know When to Act
Control charts separate signal from noise. Only special causes (points outside control limits, runs, trends) warrant adjustment. A point beyond 1 sigma is NORMAL -- it happens 32% of the time.
Tampering is Expensive
Every unnecessary adjustment costs money twice: the adjustment itself AND the additional variation it creates. The over-adjuster's total cost was much higher despite trying harder.
Trust Your Process
If your process is in statistical control, the best thing you can do is leave it alone. Improvement comes from reducing common cause variation (better tooling, materials, methods) -- not from chasing individual measurements.
Try It Yourself
Want to check YOUR process capability? Paste your data into our free tool.
Open SPC Quick CheckStill have questions? Ask QC-Coach
Tap a question below or type your own
QC-Coach
AI quality coaching assistantGet personalized AI insights on your analysis. QC-Coach explains your results, identifies issues, and suggests next steps.
Found this useful? Share it with a colleague.