When ‘In Spec’ Fails: A Case Study of Variation, Collaboration, and Change

In an industry where precision and reliability are paramount, uncovering hidden defects in equipment can mean the difference between operational efficiency and costly downtime.

A global industrial equipment manufacturer was facing persistent connector failures that standard quality checks couldn’t explain. Initial assumptions pointed to supplier issues, but deeper investigation revealed a more complex picture. By applying advanced statistical tools including Type 1 and Type 2 Measurement Systems Analyses, and challenging long-held assumptions, the team uncovered hidden variation and unrepresentative sampling that had masked the true root causes. Through a structured, collaborative approach involving Gemba walks, supplier engagement, and a full connector redesign, they achieved an 82% reduction in defects, a 27% decrease in rework and line stoppages, and strengthened supplier relationships through clearer, faster feedback mechanisms. This case study demonstrates how rigorous data analysis and cross-functional collaboration can drive lasting, scalable improvements in quality and reliability.

Achievements

Defect Reduction

82% decrease in faulty connectors reaching assembly, based on pre- and post-intervention sampling data.

Early Issue Identification

54% faster detection of supplier-related issues through improved incoming inspection and clearer feedback loops.

Measurement Clarity

>90% confidence in critical dimension data post-MSA, up from <60% due to improved gauge accuracy and operator consistency.

Production Efficiency Gains

27% reduction in rework and line stoppages, improving overall throughput and resource utilisation.

When ‘In Spec’ Fails: A Case Study of Variation, Collaboration, and Change- image - 3

The Customer Challenge

The customer operates a global inventory of industrial equipment and was aware of an ongoing issue with connector failures. With strict adherence to engineering tolerances and a robust Quality Assurance (QA) programme, sampling of critical dimensions consistently returned values within specification, leaving the root cause elusive. Initial assumptions pointed to supplier quality, but the true root causes were more complex and required a structured approach to uncover.

This challenge is common across many industries where precision and reliability are crucial. Even with rigorous quality assurance measures, hidden variations and unrepresentative samples can compromise product performance and reliability, making a thorough investigation and innovative solutions essential.

When ‘In Spec’ Fails: A Case Study of Variation, Collaboration, and Change- image - 4

“We had a Quality Assurance programme in place which verified that the sampled values were within spec, so we couldn’t understand the reasons for the failure.”

The Work Completed

Gemba Walk and Process Mapping

The investigation began with a Gemba walk; an on-site observation of the connector installation process. This allowed the team to witness first-hand how the connectors were handled, assembled, and inspected.

Through detailed process mapping, they identified subtle but critical inconsistencies in operator technique, environmental conditions, and inspection routines. These inconsistencies were not visible in standard reports but had a cumulative impact on connector performance. The exercise helped isolate where defects were being introduced and highlighted gaps in process control that had previously gone unnoticed.

Measurement System Analysis (MSA)

The team turned their attention to validating the measurement system to ensure data integrity. They carried out both Type 1 and Type 2 Measurement Systems Analyses to assess the critical dimensions.

This revealed inconsistencies in how connector defects were being identified and recorded, and uncovered that the original samples were not representative of the entire population. These insights raised confidence in critical dimension data from less than 60% to over 90%, establishing a far more reliable foundation for further investigation and improvement.

Supplier Engagement and Root Cause Analysis

With internal processes better understood, the team engaged directly with suppliers to explore upstream factors. Structured root cause analysis workshops were held to examine each stage of the connector’s lifecycle, from raw material selection to final inspection.

This collaborative approach uncovered specific process failures, such as uncontrolled variation in tooling wear and undocumented handling steps. By working side-by-side with suppliers, the team was able to implement corrective actions that were both practical and sustainable, strengthening the partnership and improving accountability.

When ‘In Spec’ Fails: A Case Study of Variation, Collaboration, and Change- image - 7

The Results and Benefits the Customer Achieved

The structured, data-led approach led to an 82% reduction in faulty connectors reaching assembly, directly improving the reliability and availability of operations. Improved collaboration with suppliers enabled issues to be identified 54% faster, reducing the risk of downstream defects and delays. These improvements also enhanced production flow, with fewer delays and less rework, and helped build stronger, more transparent relationships with key suppliers.

A pivotal factor in the project’s success was the team’s commitment to challenging assumptions and digging deeper into the data. Guided by the principle, “The data is telling you a story that you don’t understand; don’t assume the data is wrong, instead assume your assumptions are wrong and work to uncover the truth,” the team applied advanced statistical methods to uncover hidden variation and measurement inconsistencies. This mindset fostered a collaborative environment between quality assurance, engineering, and suppliers, ultimately enabling a comprehensive redesign of the connector and a more resilient supply chain.

Crucially, the initiative established a sustainable framework for applying advanced statistical thinking to quality management, one that continues to deliver value across other product lines. As a team member confirmed, “We now have the tools to tackle complex problems and drive real change.”

When ‘In Spec’ Fails: A Case Study of Variation, Collaboration, and Change- image - 8

This case study is a clear reminder that:

Apparent compliance with spec doesn’t guarantee performance.
Poor measurement systems can mask real problems.
Supplier collaboration is most effective when it’s grounded in shared data and mutual learning.
Embedding the learning is as important as fixing the issue.

Do these challenges feel familiar? We’ve helped others navigate them, and we’d be happy to explore how we could support you too.

Let's Talk

Employee Engagement, Starting with Listening

MEETS US AT THE GEMBA!

GEMBAfication is PMI’s new podcast for improvement professionals looking for an honest view of what improvement really looks like in practice. Because in our world, every experience is valuable.

Listen Now