Iris Optics
← Back to Representative Work
Data PipelinesOptical MetrologyManufacturing Test

AR/VR Waveguide OQC Dashboard

Building a test-data dashboard for optical KPI monitoring, lot disposition, yield analysis, and cross-vendor comparison.

Confidentiality & Anonymization Note

These examples are generalized and anonymized to illustrate Iris Optics capabilities. Details are modified to protect confidentiality and avoid disclosure of proprietary employer, client, or product information.

Challenge

A representative optical hardware program may face high-dimensional optical KPI data spanning multiple builds, configurations, and component vendors, making Outgoing Quality Control (OQC) decisions slow and complex.

Approach

We developed a workflow to architect structured data pipelines mapping complex measurement data to specifications. We organized the data into KPI families and performed automated Cpk and yield analysis, surfacing results in targeted dashboards.

Outcome

This methodology demonstrates how engineering and quality teams can achieve significantly faster review cycles, better visibility into supplier performance, and clearer, data-driven release decisions.

Background

Advanced optical components like AR/VR waveguides generate vast amounts of measurement data during quality control. When multiple vendors are supplying different configurations, tracking performance and yield across builds becomes a massive data management challenge.

The Challenge

Engineering teams are often overwhelmed by high-dimensional optical KPI data. Different vendors provide data in varying formats, and the relationships between different optical performance metrics are not easily comparable. Making objective, rapid disposition decisions at OQC is difficult without a unified framework, often requiring hours of manual data wrangling.

The Approach

We developed a systematic framework to build structured data pipelines specifically for optical test data:

  1. Data Standardization: We designed parser scripts to ingest varying vendor data formats into a unified schema.
  2. Specification Mapping: We mapped the ingested data directly against product specification tables, automatically flagging out-of-tolerance parameters.
  3. KPI Family Grouping: We organized hundreds of individual metrics into logical “families” (e.g., efficiency, uniformity, stray light) to make the data digestible.
  4. Dashboard Implementation: We built interactive dashboard views that presented automated Cpk (process capability) calculations, yield trends, and cross-vendor comparisons.

The Outcome

This automated dashboard workflow transforms the OQC process. What previously took days of manual analysis is reduced to minutes. The quality team gains immediate visibility into supplier performance trends, enabling clearer, faster release decisions and accelerating the overall hardware development cycle.