Trainer Guide: CNC Bore SPC Simulator

Complete trainer guide for the SPC Factory Simulator CNC bore scenario. Includes correct responses, scoring rubric, teaching moments, and facilitator notes for quality training sessions.

12 min readAll Levels

Trainer Guide: "The Bore is Drifting"

The complete answer key and facilitator guide for the CNC bore tool wear simulation. Share this with consultants, trainers, and instructors to use as part of SPC training programs.

Trainers & Consultants 35-min session format Share freely

Scenario Overview

Machine	CNC Lathe, Cell 14
Part	Transmission housing bore
Characteristic	Bore inner diameter
Target	25.000 mm
USL	25.050 mm
LSL	24.950 mm
Measurement	Digital bore gauge, 0.001 mm resolution
Chart type	Individuals (X-mR)
Production rate	1 part every 5 minutes
Shift	06:00 - 14:00 (96 subgroups)
Injected event	Progressive tool wear starting at subgroup 26

The Correct Response -- Step by Step

Phase 1: Establish Baseline (Subgroups 1-25)

What the quality engineer should observe:

The process is stable and in statistical control
Mean is approximately 25.002 mm (slight positive offset from tool deflection -- normal)
Natural variation (sigma) is approximately 0.008 mm
No Western Electric rule violations; all points within specification limits

What the quality engineer should do:

Let the chart establish control limits from the first 20-25 subgroups
Note the UCL, CL, and LCL values
Do NOT stop production -- the process is in control

Training point: Many trainees stop production early because they are anxious. A false alarm wastes production time and money. The simulator penalizes this with -10 points per false alarm, teaching the real-world consequence: unnecessary stoppages cost $500-2,000 on a CNC cell.

Phase 2: Detect the Signal (Subgroups 26-37)

At subgroup 26, the carbide insert begins wearing. The bore diameter gradually increases at +0.001 mm/subgroup -- the classic tool wear trend.

26-32

Subtle drift

Upward drift forming. Points may not look alarming individually, but consecutive points above CL are accumulating.

FIRST SIGNAL (Rule 2)

9 consecutive points above the center line. An alert quality engineer should stop here.

OBVIOUS (Rule 1)

First point above UCL. Anyone watching the chart should stop production immediately.

Optimal response: Stop production at subgroups 33-35. Stopping at subgroup 33 earns maximum detection points (40/40). Each additional subgroup after 35 costs -5 points.

Training point: The difference between a skilled quality engineer and a novice is catching the Rule 2 violation (trend/run) BEFORE the Rule 1 violation (point beyond limits). Rule 1 is obvious; Rule 2 requires pattern recognition. This is the core skill the simulator teaches.

Phase 3: Stop Production and Investigate

After stopping, the quality engineer should immediately:

Stop the machine (hit the red STOP button)
Quarantine suspect parts -- everything from subgroup 26 onward
Review the 5 factory logs to identify the root cause

What to look for in each log:

Tooling Log ROOT CAUSE

Insert #VT-2247 installed at shift start; tool life counter at 165-170 / 200 parts (83-85%)
Automated alert triggered at 80% life recommending a scheduled change
No action was taken on the alert

The carbide insert is approaching end of life. As the cutting edge wears, it deflects more under cutting forces, producing progressively larger bore diameters.

Raw Material Log RED HERRING

New material lot #3012-B from a new supplier received at 07:10. Hardness passed (29 HRC within 26-32 spec).

Not the cause: timing does not correlate (trend starts at subgroup 26, ~08:10). Material changes cause sudden shifts, not gradual trends.

Operator Log RED HERRING

Operator took a 15-minute break at subgroup 16. Machine ran unattended in auto-cycle.

Not the cause: trend starts 10 subgroups after the break. An unattended CNC in auto-cycle does not change process parameters.

Environment Log RED HERRING

Shop temperature rose 1.5-2 deg C over the shift, with a brief spike when the loading dock door opened.

Not the cause: thermal expansion from 2 deg C would produce only ~0.001-0.002 mm drift -- an order of magnitude less than the 0.015+ mm observed.

Maintenance Log SUPPORTING

All PM checks passed at shift start. No maintenance alerts were triggered. The absence of a proactive tool change alert confirms a gap in the tool management process.

Phase 4: Annotate the Chart

Field	Correct Answer
Subgroup	26 (+/- 2 for full credit, i.e., 24-28)
Annotation type	Tool wear
Note (example)	"Tool wear -- carbide insert at 85% life, trend started here. Insert change required."

Training point: Chart annotation is critical but often neglected. Every special cause event should be documented on the chart with WHAT happened, WHEN it started, and WHAT WAS DONE. This creates institutional knowledge that future operators and engineers can reference.

Phase 5: Corrective Action (Discussion Topic)

In a real factory, the quality engineer would also need to take these four levels of corrective action:

Immediate

Change the carbide insert and verify the first-piece dimension returns to target.

Containment

Sort suspect parts (subgroups 26+). Measure each bore. In-spec ships; out-of-spec is scrapped or reworked.

Preventive

Reduce tool life limit from 200 to 170 parts, or implement offset adjustment at 80% life.

Systemic

Review why the 80% alert was not acted upon. Add a mandatory stop-and-verify checkpoint to operator work instructions.

Scoring Rubric

Category	Max	What it measures	Real-world skill
Detection Speed	40	How quickly you stopped after the first signal	Pattern recognition on control charts
Root Cause ID	30	Correctly identified tool wear	Systematic investigation using factory data
Annotation	20	Correct subgroup and type on the chart	Documentation discipline
Parts Saved	10	How few bad parts shipped	Economic awareness of quality failures

90-100

Master SPC Analyst

70-89

Solid Quality Engineer

50-69

Learning Technician

< 50

Trainee

Key Teaching Moments

The single most important lesson: detecting trends and runs (Rule 2) before points exceed control limits (Rule 1) is what separates proactive from reactive quality control. Rule 1 violations mean the damage is already done.

Multiple logs contain plausible-looking events (new material lot, temperature rise, operator break). Students must learn to correlate TIMING of chart signals with TIMING of log entries. The trend starts at subgroup 26 (~08:10). Only events at or just before that time are candidates.

Chart Pattern	Likely Cause Type
Gradual upward trend	Tool wear, thermal drift
Sudden step change	Material lot, operator change, setup adjustment
Increasing spread	Fixture/workholding, bearing wear, material inconsistency
Cyclic pattern	Mechanical component (bearing, gear), batch processing
Single outlier then return	Measurement error, momentary disturbance

Each subgroup of delay represents 5 minutes and potentially one more out-of-spec part. Optimal detection (subgroup 33): 0 bad parts. Late detection (subgroup 38): 4+ bad parts, -20 points. Failure (subgroup 50): 8+ bad parts, customer complaint. In a real factory producing $50 parts with $5,000 complaint costs, the difference between subgroup 33 and 50 detection is $850+ in scrap plus $5,000+ in complaint investigation.

Unannotated control charts are just data. Annotated charts are knowledge. When the next shift engineer sees a flag on subgroup 26 reading "Tool wear -- insert changed at 170 parts, new limit set to 160 parts," they know what happened, what was done, and what the new operating parameter is. This is how manufacturing organizations learn.

Facilitator Notes

Before the Session

Walk through the briefing screen. Ensure everyone understands the X-mR chart concept.
Explain the simulation runs in real-time -- they need to watch actively.
Mention false alarms are penalized. Don't stop the line without evidence.

During the Session

Let participants run independently (phone or laptop).
Do NOT reveal the root cause in advance.
Encourage tapping data points to see timestamps for log correlation.
In groups: discuss chart observations before revealing logs.

After the Session

Compare scores. Discuss who caught it earliest and what they saw.
Review wrong root cause selections. What led them astray?
Use score breakdowns to identify which skills need reinforcement.

Suggested 35-Minute Session Format

Time	Activity
5 min	Intro: What is SPC? Why control charts?
5 min	Demo: Walk through the briefing screen and controls
10 min	Play: Participants run the simulation
10 min	Discussion: Compare results, discuss root cause
5 min	Debrief: Key takeaways, connect to their factory
35 min	Total

Frequently Asked Questions from Trainees

The chart DID catch it at subgroup 33 via Rule 2 (9 consecutive points above CL). The question is whether the quality engineer was watching closely enough. Rule 2 violations are subtle -- the individual points are all within limits. It takes discipline to monitor run patterns, not just limit violations.

Automated SPC alerts are valuable, but they create alert fatigue risk. Operators who receive 10 false alerts per shift start ignoring them. This simulator develops the human judgment that complements automated systems. A skilled quality engineer knows when an alert is real and when it can wait.

Yes, but you would prioritize. Check the 5M+E categories (Man, Machine, Material, Method, Measurement, Environment) and ask: "What changed at the time the signal started?" This focused approach narrows the investigation to 1-2 logs quickly.

Same analysis, different direction. A downward trend on bore diameter could indicate tool buildup (material welding to the cutting edge) or thermal contraction. The investigation methodology is identical: correlate chart timing with log entries, identify the assignable cause, annotate, and correct.

For practical factory use, focus on Rules 1 and 2. Rule 1: Any point beyond 3-sigma (the control limits) -- obvious, immediate action. Rule 2: 9 consecutive points on one side of the center line -- subtle, requires attention. These two catch the vast majority of real-world process disturbances.

Ready to Try It?

Run the simulation yourself, or share the link with your training group.

Launch the Simulator

Try It Yourself

Want to check YOUR process capability? Paste your data into our free tool.

Open SPC Quick Check

Still have questions? Ask QC-Coach

Tap a question below or type your own

QC-Coach

AI quality coaching assistant

Get 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.