Supplier Corrective Action (CAR) Case Study Analysis: A Complete Closed Loop from 8D Root Cause to Recurrence Prevention

1. A Typical Incoming Quality Incident

In the fall of 2025, a motor assembly line at a certain automotive parts company (hereinafter referred to as "the company") experienced batch stoppages for two consecutive days: the sealing rings provided by a supplier fell off after being pressed, causing the air-tightness test failure rate to surge from the usual 0.2% to 6.8%. Over the two days, more than 1,200 nonconforming products were produced, resulting in direct losses that included rework labor, production stoppage, and emergency freight costs, totaling over 400,000 yuan.

The quality department immediately conducted an investigation and confirmed that the issue was concentrated in two batches from the same supplier. According to the company's supplier quality management procedure, such a "batch nonconformity plus production line stoppage" event is classified as the highest level of quality incident, necessitating the initiation of a Corrective Action Request (CAR) and requiring the supplier to provide temporary containment measures within 48 hours and complete a root cause analysis within 15 days. A deep investigation into "why the sealing rings fell off" thus began.

It is important to note that the background of this incident was not unusual. The supplier had been a long-term partner of the company for over five years, with a consistent incoming inspection pass rate of over 99%. Neither party had anticipated such a concentrated batch failure. It was precisely this "long-term stability" that led to complacency, allowing the issue to fester for several weeks before it erupted. A post-incident review revealed that the drift in formula and process parameters had been evident for a month prior to the incident, with the supplier's internal process capability index (Cpk) gradually declining from 1.3 to 0.9. However, due to a lack of trend monitoring, these signals were dismissed as normal fluctuations. This incident serves as a reminder that the focus of incoming quality management should not only be on maintaining batch pass rates but also on monitoring process trends for anomalies.

2. Temporary Containment: Stop the Bleeding Before Treating the Wound

The first step after issuing a CAR is not to assign blame but to contain the issue—preventing nonconforming products from continuing to flow into the production line. Within 24 hours of receiving the CAR, the supplier implemented three temporary measures: first, isolating all sealing rings in the finished goods warehouse and those in transit, involving eight batches totaling about 60,000 pieces; second, adding a retest for air-tightness on motors that had been pressed but not yet shipped, ensuring that the problematic batches did not leave the factory; third, dispatching a dedicated team to assist the company in screening and replacing all sealing rings from the same batch in inventory with conforming batches.

All three measures were fully implemented within 72 hours, and the production line resumed normal operations. The value of containment lies in minimizing losses—without this step, the problematic batches might have continued to enter production, leading to a significant increase in stoppage time and rework costs. The first principle of the CAR process is "isolate the risk before analyzing the cause."

A crucial detail often overlooked is that containment measures must be documented and traceable. The company specifically included a "containment evidence" section in the CAR form, requiring the supplier to attach photos of the isolation area, screening records, and retest reports, rather than just an oral assurance that the issue had been handled. This requirement played a critical role in subsequent verification—thanks to the complete containment records, both parties were able to accurately define the loss scope within two weeks and rule out the interference variable of "mixed nonconforming products." Containment is not just a formality but the first cornerstone of the entire CAR closed loop.

3. Root Cause Analysis: 5Why Questions Penetrate Three Layers of Causes

After containment was completed, the core phase of the CAR process—root cause analysis—began. The supplier's initial report attributed the cause to "operator negligence, misalignment during pressing." The company's quality engineer did not accept this answer and required the supplier to continue the 5Why method and verify on-site.

First question: Why did the sealing rings fall off? — There was approximately a 0.3 mm gap between the sealing ring and the end cover groove after pressing. Second question: Why was there a gap? — The outer diameter of the sealing ring was too small, exceeding the lower tolerance limit on the drawing. Third question: Why was the outer diameter too small? — The shrinkage after vulcanization was about 0.5% larger than the design value. Fourth question: Why did the shrinkage increase? — The amount of vulcanizing agent in the formula deviated from the standard process, and the vulcanization temperature was below the lower limit of the process window. Fifth question: Why did the formula and temperature deviate? — The workshop temporarily adjusted the shift schedule to meet the delivery deadline, and the new team did not verify the formula weighing records according to the work instruction, nor did they check the temperature controller before the shift.

After five layers of questioning, the truth emerged: what appeared to be "operator negligence" was actually a failure at three levels by the supplier—non-compliance with process discipline (not following the work instruction), lack of process monitoring (no temperature controller inspection, no poka-yoke), and absence of management mechanisms (no change management during rush delivery, no first article inspection).

It is essential to emphasize that each layer of the 5Why conclusion must be supported by evidence, not just inference. The company's quality engineer added two verification steps: first, retrieving the supplier's formula weighing records and temperature control curves for the shift, confirming that the amount of vulcanizing agent and the vulcanization temperature indeed deviated; second, conducting a sampling inspection on the finished products in inventory from the same formula and batch, confirming that the issue was limited to these two batches and did not affect other products. These two verification steps were crucial for confirming the validity of the cause and the controllability of the scope. Thus, the root cause was traced from the "human" level to the "system" level, laying a solid foundation for subsequent measures.

4. Corrective and Preventive Actions: From Symptomatic Treatment to Fundamental Solutions

Based on the root cause analysis, the supplier formulated five measures, divided into corrective and preventive levels.

Corrective actions (addressing the occurred issue):

1. Revising the work instruction to add a dual-person verification and photo documentation step in the formula weighing process.
2. Incorporating the vulcanization temperature into the equipment's automatic recording system, with immediate alarms and shutdowns if it exceeds the process window.
3. Restoring the pre-shift inspection system and shortening the temperature controller calibration cycle from six months to three months.

Preventive actions (preventing similar issues from recurring):

1. Installing a poka-yoke device, linking the weighing process with the formula database, and locking the machine if the weighing deviation exceeds the set value.
2. Establishing a first article inspection system, requiring the first piece to be inspected and approved after each shift change before batch production can proceed.
3. Incorporating "rush delivery temporary adjustments" into the change management process, ensuring that any changes are confirmed by both process and quality departments.

The company's quality engineer reviewed each measure to ensure its relevance—each measure must address a specific failure point. The introduction of the poka-yoke device, in particular, upgrades "human verification" to "system interception," which is key to preventing recurrence. During the review, two principles were established: first, corrective actions must target systemic failures confirmed in the root cause analysis, and any measures addressing only surface issues will be rejected; second, the completion standards for each measure must be verifiable, such as having signed records for "dual-person verification" and alarm logs for "temperature alarms," to ensure that measures are not just停留在纸面上 (stuck on paper).

5. Effect Verification and Horizontal Deployment

After the measures were implemented, the company did not rush to close the CAR but instead set a three-month observation period. The verification data showed: in the first month, the incoming batch pass rate for sealing rings recovered to over 99.8%; in the second month, the air-tightness test failure rate dropped to 0.1%, lower than pre-incident levels; in the third month, the supplier achieved zero defects in twelve consecutive batches. Additionally, the company compared the process capability index (Cpk) before and after the rectification, which improved from 0.83 to 1.42, confirming that the process was truly under control.

More importantly, the horizontal deployment was critical. The company required the supplier to apply the lessons learned from this CAR to all rubber product lines: the same poka-yoke weighing device was installed on two other production lines, and the temperature alarm logic was incorporated into the group standards. Three months later, the supplier also achieved zero defects in similar issues for other products. The company recorded this incident in the supplier quality knowledge base, listing it as a key inspection item for the annual supplier audit, and reported it to all suppliers at the quarterly meeting to promote a broader understanding and application of the lessons learned.

The value of horizontal deployment is often underestimated. Many companies treat CARs as one-off tasks, closing the issue and moving on, only to see similar problems reappear at another supplier or on another production line. The company's approach is to answer three questions after each CAR closure: Is this issue possible at other suppliers? Are there corresponding interception methods in our inspection standards? Is there a need to upgrade the drawings, specifications, or control plan? This "three-pronged" mechanism effectively helps suppliers move beyond the "addressing the issue at hand" mindset.

6. CAR Trigger Conditions and Tiered Response Mechanisms

This case can lead to a practical question: What types of incoming quality anomalies require a CAR? Not all nonconformities necessitate a CAR, as overuse can render the mechanism ineffective. Reasonable trigger conditions typically include four categories:

1. Repeated issues, where the same type of defect from the same supplier appears twice or more within three months, indicating systemic process gaps.
2. Severe quality incidents, such as critical defects, batch nonconformities (batch pass rate below the agreed AQL limit), or material anomalies causing production line stoppages.
3. Nonconformities discovered during audits, whether second-party or third-party, should be tracked and rectified through CARs.
4. Supplier responsibility issues traced from customer complaints.

The company's approach is to categorize CARs into three levels:

• General level: the supplier must respond with a rectification plan within 10 working days.
• Severe level: the supplier must respond within 5 working days.
• Emergency level (major quality incident): the supplier must provide temporary containment measures within 48 hours and complete the root cause analysis within 15 days, with the supplier's senior management signing off on it.

The significance of tiered responses is to allocate limited review resources to the highest-risk issues and to make it clear to suppliers which issues will be escalated. Additionally, the problem description in the CAR form must be objective and quantifiable, such as "sealing ring outer diameter is 0.3 mm smaller, exceeding the lower tolerance limit on the drawing," rather than vague statements like "poor quality" or "incorrect dimensions." The more precise the description, the smoother the subsequent root cause analysis and responsibility determination will be.

7. Post-incident Review: Three Key Takeaways for the Company

1. The value of a CAR lies in root cause analysis, not in claims. If the supplier's initial cause of "operator negligence" had been accepted, the issue would have inevitably recurred. What truly resolved the problem was the five-layer questioning that penetrated to the process and system levels, coupled with the introduction of poka-yoke devices. When reviewing the supplier's root cause analysis, the company must insist on "three questions": Why did the self-inspection not intercept the issue? Why were similar issues not found in other products? Where did the quality management system fail? These three questions effectively help suppliers break out of the "addressing the issue at hand" mindset.

2. Containment and verification are both essential. Temporary containment determines the upper limit of losses, while effect verification confirms the authenticity of improvements. Closing a CAR without verification is equivalent to handing the problem back to chance. Using a three-month observation period and Cpk data to speak, the company can confirm that the improvements have truly been implemented. During the verification stage, particular attention must be paid to "whether the sample size is sufficient and whether the observation period covers seasonal variations" to avoid drawing overly optimistic conclusions from short-term data.

3. Recurrence prevention relies on the system, not individuals. From dual-person verification, automatic alarms to poka-yoke lockout, the essence of this comprehensive approach is to transform "not making a mistake this time" into "making it difficult to make a mistake ever again." When corrective actions touch on poka-yoke and process design, the supplier's quality capability truly improves, and the supply chain's resilience is enhanced. A complete closed loop of a CAR is essentially a quality capability upgrade driven by root cause analysis in the supply chain.

For quality managers, every incoming quality anomaly is a rare opportunity for improvement: it exposes cracks in daily management that are otherwise invisible and points to the direction for system upgrades. Transforming CARs from a "penalty tool" to a "capability building tool" ensures that suppliers and buyers truly stand on the same boat.

A complete closed loop of a CAR is essentially a quality capability upgrade driven by root cause analysis in the supply chain.

Knowledge code: 9.2.3

Version: v20260801

Author: Quality Think Tank Quality Think Tank is dedicated to providing systematic professional knowledge, methodologies, and practical tools for quality management practitioners, helping companies continuously improve their quality capabilities.