Case Study: How an Automotive Supplier Reduced Brake Component Defect Rate from 3,200 to 180 PPM Using DMAIC

Learn how a Tier-1 automotive supplier achieved 94% defect reduction (from 2 Sigma to 6 Sigma) using Six Sigma DMAIC methodology, preventing $2.8M in annual warranty claims and customer penalties while improving OEE from 76% to 91%.

Executive Summary

Organization: Tier-1 automotive supplier with $850M annual revenue, 4 plants globally, supplying brake components to three major OEMs (GM, Ford, Stellantis)

Challenge: Brake caliper assembly line producing 3,200 PPM (0.32%) defect rate, primarily driven by porosity in aluminum castings (58% of defects) and dimensional non-conformance (42%). Customer (OEM) threatened to cancel $45M annual contract if defect rate was not reduced to under 500 PPM within 6 months.

Solution: Six Sigma DMAIC project implementing three key improvements: induction furnace temperature control upgrade, predictive tool management system with in-process measurement, and standardized operator training and certification program.

Results: Defect rate reduced from 3,200 PPM to 180 PPM (94% reduction), sigma level improved from 2.1σ to 6.1σ, process Cpk improved from 0.51 to 1.87, OEE increased from 76% to 91%, with total annual financial impact of $62M+ and 28:1 ROI.

Business Problem and Impact

The brake caliper assembly line at Plant 2, Line 3A was experiencing severe quality issues with a defect rate of 3,200 PPM, placing the process at 2.1 Sigma level — well below automotive industry standards and customer requirements. This created multiple business risks:

• Warranty Claims: $1.8M annually from field failures and product recalls
• Scrap and Rework: $680K in internal quality failures
• Customer Penalties: $340K in PPM non-conformance fees
• Poor OEE: 76% Overall Equipment Effectiveness, 10% below industry benchmark due to quality stops
• Contract at Risk: $45M annual revenue threatened by customer cancellation notice
• Low Process Capability: Cpk of 0.51 indicating severely incapable process

The defects were concentrated in three Critical-to-Quality (CTQ) characteristics: casting porosity (1,856 PPM, Cpk=0.42), bore diameter (944 PPM, Cpk=0.89), and surface finish (400 PPM, Cpk=1.12).

DMAIC Methodology Application

Define Phase: Project Charter and CTQ Definition

Problem Statement: Brake caliper assembly line (Plant 2, Line 3A) produces 3,200 PPM defect rate, driven by casting porosity (58%) and dimensional non-conformance (42%). Current sigma level 2.1σ. Customer requires less than 500 PPM or will cancel $45M contract.

Goal Statement: Reduce brake caliper defect rate from 3,200 PPM to under 500 PPM (84%+ reduction) by Q2 2024, achieving minimum 4.5σ level, while maintaining production throughput of 2,400 units per shift.

Three CTQ characteristics were defined with specifications:

• Casting Porosity (X-ray): ≤Grade 2 per ASTM E505 standard — Current: Cpk=0.42, 1,856 PPM
• Bore Diameter: 54.00mm ± 0.05mm — Current: Cpk=0.89, 944 PPM
• Surface Finish (Ra): ≤1.6 μm — Current: Cpk=1.12, 400 PPM

Measure Phase: Measurement System Analysis and Baseline

Measurement System Validation: Before collecting baseline data, the team validated all measurement systems using Gage R&R studies:

• Porosity (X-ray): Attribute Gage R&R with 3 inspectors, 30 samples, 3 repetitions — Results: 94% repeatability, 91% reproducibility, 96% accuracy vs reference (MSA acceptable)
• Bore Diameter (CMM): Continuous Gage R&R with 10 parts, 3 operators, 3 repetitions — Results: 8.2% R&R, 9.4% tolerance, NDC=8 (MSA acceptable)
• Surface Finish: Continuous Gage R&R — Results: 12.3% R&R, 18.1% tolerance, NDC=6 (MSA marginally acceptable)

Baseline Data Collection: 5,000 consecutive parts were measured over 2 shifts × 5 days production. Control chart analysis revealed the process was NOT in statistical control, with 18 out-of-control signals detected in 25 subgroups (Rule 1, 2, and 3 violations indicating special causes were present).

Process capability for the worst CTQ (casting porosity) showed Cpk=0.42, defect rate of 1,856 PPM, and sigma level of only 1.8σ. The estimated Cost of Poor Quality (COPQ) for porosity defects alone was $1.2M annually.

Analyze Phase: Root Cause Identification

Using fishbone diagrams, statistical analysis, and process investigation, the team identified three primary root causes:

1. Casting Process Temperature Variation (Root Cause for Porosity)

Finding: Molten aluminum temperature varied ±15°C (specification: ±5°C). Statistical analysis showed temperatures below 700°C correlated with 4.2× higher porosity rate (p=0.001).

Root Cause: Induction furnace temperature controller was failing. PID loop tuning had degraded over 3 years with no preventive calibration schedule in place.

2. CNC Tool Wear Pattern (Root Cause for Bore Diameter Drift)

Finding: Bore diameter drifted +0.035mm over tool life (approximately 800 parts). Tool changes were reactive at failure rather than proactive based on data.

Root Cause: No tool life monitoring system existed. Operators used visual inspection ("eyeball") to determine tool condition. Tool change intervals were based on operator "feel" rather than objective data.

3. Operator Training Gaps (Contributing Factor)

Finding: Defect rate was 2.1× higher on night shift compared to day shift (p=0.008). Night shift operators averaged only 8 months experience versus 24 months for day shift.

Root Cause: No standardized training program. Night shift coverage relied on temporary and contract workers with minimal training and no certification requirements.

Improve Phase: Solution Implementation

The team implemented three interconnected solutions in a phased approach:

Solution 1: Furnace Temperature Control Upgrade

Implementation: Replaced failing temperature controller, implemented advanced PID tuning with automatic adaptive control, and added real-time SPC monitoring with automated alerts at ±8°C deviation.

Results: Temperature variation reduced from ±15°C to ±3°C, porosity defects decreased 78% (from 1,856 to 408 PPM), and Cpk improved from 0.42 to 1.31.

Investment: $42K including controller hardware, installation, and tuning services.

Solution 2: Predictive Tool Management System

Implementation: Installed in-process measurement (IPM) system monitoring bore diameter every 50 parts with automated tool change trigger at +0.015mm drift. Built comprehensive tool life database to optimize change intervals.

Results: Bore diameter variation reduced 68% (σ decreased from 0.022mm to 0.007mm), defect rate decreased 90% (from 944 to 95 PPM), Cpk improved from 0.89 to 1.89, and tool life increased 23% through optimized change point.

Investment: $125K including IPM system hardware and software integration.

Solution 3: Standardized Training and Certification Program

Implementation: Developed comprehensive 40-hour certification program combining classroom instruction and on-the-job training. Made certification mandatory for all operators with skill matrix tracking and prohibition of solo operation until certified.

Results: Night shift defect rate reduced 58% (eliminating shift effect), setup time reduced 35% due to better trained operators, and operator confidence scores improved 67% based on surveys.

Investment: $38K including curriculum development and paid training time.

Control Phase: Sustainment and Monitoring

A comprehensive control plan was established with different monitoring approaches for each CTQ:

• Casting Porosity: Target Cpk ≥1.33, 100% X-ray inspection, P-chart hourly monitoring, furnace temperature SPC with automatic alerts
• Bore Diameter: Target Cpk ≥1.67, IPM measurement every 50 parts, X-bar-R chart with real-time display and automated tool change triggers
• Surface Finish: Target Cpk ≥1.33, one sample per hour measurement, I-MR chart at shift level with trending analysis

After 18 months post-implementation, performance remained sustained with average defect rate of 165 PPM (range: 120-210 PPM, process in statistical control), sigma level of 6.2σ maintained, OEE sustained at 91% (15-point improvement maintained), zero customer penalties for 18 consecutive months, and the $45M contract successfully renewed with an additional $12M expansion order.

Financial Results and ROI

Cost Savings (Annual):

• Warranty claims: $1.68M (93% reduction from $1.8M baseline)
• Scrap and rework: $612K (90% reduction from $680K baseline)
• Customer penalties: $340K (completely eliminated)
• Quality labor: $172K (freed capacity for value-added work)

Revenue Protection and Growth (Annual):

• Contract secured: $45M (threatened cancellation prevented)
• Expansion order won: $12M (new business from quality demonstration)
• Premium pricing unlocked: $2.1M (quality performance bonus from customer)

Total Annual Impact: $62M+ with 28:1 ROI based on total project investment of $205K (including equipment, training, and Black Belt time).

Critical Success Factors

Measurement System Validation: Investing time upfront to validate all measurement systems through Gage R&R studies ensured data reliability. This prevented the common mistake of attempting to improve a process using unreliable measurement data.

Multi-Pronged Approach: Rather than implementing a single solution, the team addressed all three root causes simultaneously. This comprehensive approach was necessary because the root causes were independent — fixing only one would not have achieved the dramatic results.

Technology and People Balance: The project balanced technology investments (furnace controller, IPM system) with people development (training program). Both were essential — technology alone would have failed without capable operators.

Lessons Learned

Statistical Process Control Foundation: The initial finding that the process was not in statistical control (18 violations in 25 subgroups) highlighted the importance of first stabilizing a process before attempting to improve capability. Special causes must be eliminated before calculating reliable process capability indices.

In-Process vs End-of-Line Inspection: The shift from end-of-line inspection to in-process measurement (IPM every 50 parts) prevented defects rather than just detecting them. This reduced scrap, improved feedback speed, and enabled predictive tool management.

Operator Training ROI: The $38K training investment had the highest ROI of all solutions when considering soft benefits. Better trained operators reduced setup time, improved problem-solving, decreased turnover, and created a culture of quality ownership.

How This Applies to Your Organization

This case study demonstrates several universal principles applicable across manufacturing industries:

• Start with MSA: Always validate your measurement system before collecting baseline data or making improvement decisions
• Address Root Causes: Use data-driven analysis (hypothesis testing, p-values) to identify root causes rather than treating symptoms
• Balance Technology and People: Both equipment upgrades and operator development are typically needed for sustainable improvement
• Monitor and Control: Implement real-time SPC with automated alerts to sustain gains and prevent regression
• Document Everything: Comprehensive documentation enables knowledge transfer and prevents loss of learning during staff turnover

The journey from 2 Sigma to 6 Sigma is achievable with disciplined application of DMAIC methodology, appropriate technology investment, and commitment to operator development and empowerment.