Manufacturing is one of the highest-impact domains for AI — small improvements in yield, uptime, or cycle time compound across millions of units. AI in manufacturing works across the full value chain: equipment, quality, supply chain, safety, and planning.

1. Predictive Maintenance

Failure Mode Analysis

Prompt: Help me develop a predictive maintenance model for this equipment.

Equipment: CNC milling machine (Fanuc series)
Current maintenance: Time-based, every 500 hours regardless of condition
Available sensor data:
- Spindle current draw (every 5 seconds)
- Vibration (X/Y/Z axes, every 1 second)
- Temperature (spindle bearing, coolant, ambient)
- Tool wear counter
- Feed rate and depth of cut parameters

Common failure modes we've experienced:
- Spindle bearing failure (highest cost, 4 failures last year)
- Tool breakage mid-cut (causes scrap)
- Coolant pump failure
- Axis servo motor fault

Develop:
1. Key leading indicators for each failure mode
2. Monitoring thresholds and alert logic
3. Data collection architecture recommendation
4. Recommended prediction window (how far ahead to alert)
5. Model approach for each failure mode (rule-based vs ML)
6. How to measure model performance

Maintenance Work Order Language

Prompt: Convert this sensor anomaly into a maintenance work order.

Equipment: Injection molding machine #7
Anomaly detected: Hydraulic pressure trending down over 72 hours
Normal range: 180-195 bar
Current reading: 162 bar
Rate of decline: ~4.5 bar per 24 hours
Production schedule: Machine is in continuous production, next scheduled downtime is in 5 days

Write a maintenance work order:
1. Priority level with justification
2. Work description (specific checks to perform)
3. Required parts (likely causes and parts needed)
4. Estimated labor hours
5. Safety precautions
6. Decision tree: if X found, do Y
7. Recommendation: can we wait 5 days or need immediate intervention?

2. Quality Control and Inspection

Defect Classification and Root Cause

Prompt: Help me analyze this quality data to identify root causes.

Product: Aluminum die casting (automotive bracket)
Defect rate this month: 4.2% (target: <1.5%)
Defect types found:
- Porosity: 45% of defects
- Cold shuts: 30% of defects
- Flash: 15% of defects
- Misruns: 10% of defects

Process parameters collected at each shot:
- Die temperature (cavity and core)
- Injection pressure (peak and hold)
- Injection speed (first and second stage)
- Metal temperature
- Cooling time
- Shot weight

I've noticed the defect rate is highest on second shift.

Provide:
1. Likely root causes for each defect type (prioritized)
2. Process parameters most correlated with each defect
3. Hypothesis about why second shift has higher defects
4. Recommended DOE (design of experiments) to isolate root cause
5. Short-term containment actions
6. Permanent corrective action checklist

Inspection Criteria Development

Prompt: Create visual inspection criteria for a machined component.

Part: Stainless steel surgical instrument (laparoscopic grasper jaw)
Material: 17-4 PH stainless steel
Critical function: Grips tissue during surgery — must not have edges that could damage tissue or harbor bacteria
Regulatory: Class II medical device, FDA 21 CFR Part 820

Develop inspection criteria for:
1. Surface finish (Ra values, acceptable vs. reject)
2. Edge condition (burrs, sharp edges, radii)
3. Visual defects (scratches, pits, corrosion)
4. Dimensional check points (critical dimensions to measure)
5. Cleaning verification

For each criterion:
- Accept standard (what passes)
- Reject standard (what fails)
- Borderline: re-inspect or escalate
- Measurement method or reference standard
- Documentation requirement

Format: Inspection traveler table, ready to implement

SPC Chart Interpretation

Prompt: Interpret these SPC control chart data and recommend actions.

Process: Bottle fill weight (beverage production)
Target: 500g
USL: 510g, LSL: 490g
Control limits (calculated from historical data):
- UCL: 507.2g, LCL: 492.8g

Current data (last 25 samples):
[Provide your data points]

Analyze for:
1. Any Western Electric rules violations (list each violation)
2. Cp and Cpk calculation from this data
3. Process capability interpretation (capable/marginally capable/incapable)
4. Special cause vs. common cause variation present
5. Recommended actions based on findings:
   - If special cause: investigate and eliminate
   - If process drift: adjustment procedure
   - If capability issue: longer-term improvement plan
6. When to recalculate control limits

3. Production Planning and Scheduling

Capacity Planning

Prompt: Help me create a production plan for next month.

Facility: Metal fabrication shop (job shop environment)
Key resources:
- Laser cutter (1 machine, 16 hours/day, 5 days/week)
- Press brake (2 machines, 16 hours/day)
- Welding stations (8 stations, 8 hours/day)
- Painting booth (1, 24 hours/day with 4-hour cure cycle)

Open orders for next month:
Order 1: 500 units, 0.8 hrs laser / 0.5 hrs bend / 2 hrs weld / 1 hr paint
Order 2: 200 units, 0.3 hrs laser / 0.8 hrs bend / 0 weld / 0.5 hrs paint
Order 3: 300 units, 1.2 hrs laser / 1.0 hrs bend / 3 hrs weld / 2 hrs paint
[Additional orders...]

Due dates: [provide]
Setup times: 45 min laser, 30 min brake, 15 min weld station

Generate:
1. Resource loading by week (hours required vs. available)
2. Constraint identification (which resources are bottlenecks)
3. Sequencing recommendations to minimize changeovers
4. Risk assessment (where are we likely to miss due dates)
5. Options if overloaded (OT, subcontracting, due date negotiation)

Downtime Analysis

Prompt: Analyze this downtime data and prioritize improvement actions.

Plant: Automotive stamping facility
Reporting period: Last quarter (90 days)
Total available time: 3 machines × 2 shifts × 90 days × 8 hours = 4,320 hours

Downtime events (summarized):
- Die change / setup: 285 hours (65 events, avg 4.4 hrs each)
- Mechanical breakdown: 142 hours (28 events, avg 5.1 hrs each)
- Quality holds (die adjustment): 98 hours (45 events, avg 2.2 hrs)
- Material issues: 67 hours (23 events)
- Operator absence / staffing: 45 hours
- Planned maintenance: 87 hours

Calculate:
1. OEE (Overall Equipment Effectiveness) — Availability × Performance × Quality
2. Pareto chart data (by category and hours lost)
3. Cost of downtime (assume $450/hour machine rate)
4. Top 3 improvement opportunities by ROI
5. Recommended improvement approach for each:
   - SMED analysis for setup time
   - RCM analysis for mechanical failures
   - Poka-yoke for quality holds
6. Baseline metrics to track after improvements

4. Supplier Management and Procurement

Supplier Performance Review

Prompt: Write a supplier performance review letter.

Supplier: Precision Parts Co. (critical fastener supplier, 3-year relationship)
Review period: Q4 2025

Performance data:
- On-time delivery: 84% (target: 98%)
- Quality: 0.8% incoming reject rate (target: <0.3%)
- Price competitiveness: 3% above market (per recent RFQ)
- Responsiveness: Mixed — 2 unresolved corrective actions from last quarter

Significant incidents:
- November: Late delivery caused 8-hour line stoppage ($35,000 cost)
- December: Contaminated batch (debris in bag) — full lot rejected

Write a performance review that:
1. States facts objectively (no inflammatory language)
2. References specific incidents with impact
3. States expectations clearly
4. Sets corrective action requirements with deadlines
5. Defines consequences if targets not met
6. Acknowledges any positive aspects
7. Sets next review date

Tone: Firm, professional, relationship-preserving (we want to keep them if they improve)

RFQ Development

Prompt: Create an RFQ for a manufacturing component.

Component: Custom gasket (non-standard dimensions)
Material: EPDM rubber, Shore A 60-70
Dimensions: [provide]
Annual volume: 50,000 pieces
Delivery requirement: Consignment stock, 4-week supply on hand

RFQ should include:
1. Technical requirements (drawing reference, material spec, tolerances)
2. Quality requirements (PPAP level, inspection, certifications needed)
3. Packaging and labeling requirements
4. Delivery requirements
5. Pricing structure requested (tooling separate, piece price at 25K / 50K / 75K annual)
6. Sample requirements (how many, when needed)
7. Response requirements and evaluation criteria
8. Supplier qualification questions

Format: Professional RFQ document, ready to send to 5 potential suppliers

5. Safety and Compliance

Job Safety Analysis (JSA)

Prompt: Write a Job Safety Analysis for this manufacturing task.

Task: Changing grinding wheels on bench grinder
Department: Maintenance
Frequency: Weekly
Workers: 2 mechanics

Sequence of task steps:
1. Lock out / tag out power
2. Remove guard
3. Loosen arbor nut
4. Remove old wheel
5. Inspect new wheel
6. Mount new wheel
7. Replace guard
8. Test run

For each step, identify:
- Potential hazards
- Existing controls
- Additional controls needed
- Required PPE

Required regulatory references:
- OSHA 29 CFR 1910.215 (abrasive wheels)
- ANSI B7.1

Output: JSA table format, suitable for training and posting at workstation

Incident Investigation

Prompt: Help me write an incident investigation report.

Incident: Worker laceration requiring 6 stitches to left hand
Date/time: February 8, 2026, 2:15 PM
Department: Assembly
Task being performed: Trimming flash from plastic molded part with utility knife

Witness statement: Worker said blade slipped when part shifted unexpectedly. 
No cut-resistant gloves were being worn. Standard procedure requires them 
but worker said they had run out and hadn't reported it.

Investigation format:
1. Incident description (factual)
2. Sequence of events (timeline)
3. Immediate causes (unsafe act, unsafe condition)
4. Root causes (using 5-Why analysis)
5. Contributing factors
6. Corrective actions:
   - Immediate (containment)
   - Short-term (30 days)
   - Long-term (90 days)
7. Responsible parties for each corrective action
8. Follow-up verification dates

Format: OSHA recordable incident investigation report

6. Continuous Improvement

Kaizen Event Charter

Prompt: Write a Kaizen event charter.

Target area: Receiving dock / incoming inspection process
Problem: Incoming inspection takes 3 days average (target: same day)
Impact: Creates WIP, delays production schedule, requires additional warehouse space
Team: 1 process engineer (lead), 2 receiving clerks, 1 quality inspector, 1 production scheduler
Duration: 5-day Kaizen event

Charter sections:
1. Problem statement (current state, impact)
2. Scope (in scope / out of scope)
3. Goals (measurable targets for the week)
4. Team members and time commitment
5. Current state metrics to baseline before event
6. Deliverables from the event (what we will produce)
7. Sustainability plan (how we'll maintain improvements)
8. Escalation path (issues that need management decision)

Format: One-page A3 format suitable for wall posting during event

Standard Work Development

Prompt: Help me write standard work for this assembly operation.

Operation: Installing battery pack in handheld power tool
Cycle time target: 4.5 minutes
Takt time: 4.2 minutes (production rate needed)

Subtasks observed:
- Get battery pack from line-side bin (15 sec)
- Inspect battery contacts visually (10 sec)
- Slide battery into housing (20 sec)
- Click to engage latch (5 sec)
- Verify engagement (push test) (5 sec)
- Connect wiring harness (30 sec)
- Torque 4 mounting screws M4 × 8Nm (90 sec)
- Visual inspection of installation (20 sec)
- Function test (power on, check LED) (30 sec)
- Affix quality label (10 sec)

Develop:
1. Standard work chart (operations with time, walking diagram)
2. Work sequence (optimal order)
3. Standard WIP quantity
4. Quality checkpoints and error-proofing needed
5. Training points for new workers
6. Out-of-cycle work definition (what requires escalation)

Identify: Which steps are value-added vs. non-value-added

AI in manufacturing creates the highest ROI when applied to data that already exists — sensor readings, quality records, downtime logs — rather than requiring new instrumentation. Start with problems where manual analysis takes hours and data is already being collected.