Home / Case Studies / Defect Reduction & SPC

Defect Reduction & Statistical Process Control (SPC)

Process Improvement

Research-backed case study: methodology, implementation, and measurable outcomes.

Executive Summary

A critical assembly line was experiencing a high defect rate that drove rework, customer complaints, and audit scrutiny. This case study documents a structured approach: root cause analysis using the 8D (Eight Disciplines) and 5-Why methods, implementation of Statistical Process Control (SPC) with calculated control limits, real-time Power BI dashboards for monitoring and alerts, and operator training on out-of-control response. The result was approximately 40% defect reduction, faster containment via automated alerts, and audit-ready SPC documentation.

~40% defect reduction
Faster containment via automated alerts
Audit-ready SPC documentation

Context & Problem

The line in question was a high-volume assembly process where defects were discovered both in-process and at final inspection, and sometimes by the customer. Rework and scrap costs were rising, and customer quality scorecards were at risk. Management needed a data-driven response that would not only fix the immediate causes but also establish ongoing monitoring so that future shifts in the process could be detected and corrected before defects escaped.

Key challenges included: (1) multiple potential causes (material, machine, method, environment); (2) lack of real-time visibility into process stability; (3) reactive rather than preventive quality actions; and (4) need for documentation that would satisfy internal and external audits (e.g., IATF/ISO expectations around SPC and corrective action).

Methodology & Frameworks

The approach aligned with established quality frameworks and standards:

8D (Eight Disciplines) Problem Solving - Used to form a cross-functional team, define the problem, contain the issue, identify root causes, implement and verify corrective actions, and prevent recurrence. 8D is widely referenced in automotive (AIAG) and general manufacturing for serious quality issues.
5-Why - Applied within 8D to drill from symptoms to root cause (e.g., “Why did the defect occur?” through multiple layers until process or system causes were identified).
Statistical Process Control (SPC) - Control charts (e.g., X-bar/R or individual-X/moving range for key characteristics) were designed using historical data. Upper and lower control limits (UCL/LCL) were calculated using standard SPC formulas so that variation could be distinguished between common cause (inherent) and assignable cause (special).
Out-of-control response - Clear rules (e.g., points beyond limits, runs, trends) and escalation paths were defined so that operators and supervisors could react consistently when the process signaled a shift.

Implementation

After root causes were identified and short-term corrective actions were in place, SPC was rolled out for the critical characteristics that had the strongest correlation to the defects. Data was collected at the source (or pulled from existing systems) and fed into a data model. Power BI was used to build real-time control charts and a simple alert logic (e.g., flag when a point exceeded UCL/LCL or when run rules were violated). Dashboards were deployed to the floor and to quality/engineering so that everyone could see the same view of process stability.

Operator and supervisor training focused on: how to read the charts, when to stop and escalate, and how to document the response. This closed the loop between “data” and “action,” which is often where SPC initiatives fail. The SPC documentation (limits, revision history, response procedures) was organized so that auditors could easily verify that the process was monitored and that out-of-control events were addressed.

Outcomes

~40% defect reduction on the targeted assembly line over the measurement window, with a corresponding drop in rework and customer-reported issues.
Faster containment - Automated alerts in Power BI reduced the time from “process shift” to “team response,” limiting the number of nonconforming units produced.
Audit-ready SPC documentation - Control limits, basis for limits, and records of out-of-control investigations were available for internal and external audits, supporting compliance with ISO 9001 and customer-specific requirements.

Lessons Learned

(1) Root cause analysis must be disciplined (8D/5-Why) so that fixes address causes, not symptoms. (2) SPC is only effective when the organization commits to acting on out-of-control signals; otherwise it becomes “wall art.” (3) Real-time dashboards (Power BI) made the data visible and actionable at the right level. (4) Training and clear ownership for response are as important as the statistics. (5) Keeping documentation in one place and linked to the QMS simplified audit readiness.