1 / 19

Data & Analytics Applied to Test Strategy

Data & Analytics Applied to Test Strategy. Brad Waggle. Board Test Workshop. September 9th, 2014. Here in Texas, everything’s big, so we just call it data. Michael Dell … http://www.dell.com/learn/us/en/19/power/ps2q14-20140230-michael. Agenda. Context Data usage behavior shift

Download Presentation

Data & Analytics Applied to Test Strategy

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Data & AnalyticsApplied to Test Strategy Brad Waggle Board Test Workshop September 9th, 2014

  2. Here in Texas, everything’s big, so we just call it data. Michael Dell … http://www.dell.com/learn/us/en/19/power/ps2q14-20140230-michael

  3. Agenda • Context • Data usage behavior shift • Facilitating the shift • Technology Stack • What can we do now that we couldn’t do before? • What can’t we do yet but are working on?

  4. Context • Test / Supply Chain Complexity • 1000s of testers, 10’s-of-thousands of product IDs, 100’s of thousands of component part numbers • Billions and Billions of rows of data (over years and years) • Test and Repair • 6.1B • Measurement • 45.5B • Component • 65.8B • And GROWING FAST!

  5. Data usage behavior shift • In the past • Small product sample • Short time frame • Solve point problems • Lack of learning • Going Forward • All products (n = all) • All time frames • Solve strategic problems • Learn and prevent problems

  6. Facilitating the shift • Migrating from: • Excel based aggregation / manipulation • Limited custom web pages • IT driven support model • Traditional database architecture • To: • Best in class front-end analytics (Spotfire) • User driven support model / self-service • Big data architecture Test Repair Qualification Field Product Cost Component

  7. Facilitating the shift • Migrating from: • Excel based aggregation / manipulation • Limited custom web pages • IT driven support model • Traditional database architecture • To: • Best in class front-end analytics (Spotfire) • User driven support model / self-service • Big data architecture Test Repair Qualification Field Product Cost Component

  8. Technology Stack • Big Data • Cloudera Hadoop • Hive / Impala • Traditional database • MSSQL • Spotfire Analytics • R Statistical Programming (TERR)

  9. What can we do that we couldn’t do before?

  10. Use big data and analytics to: • Detect technology & quality trends • Consistently apply strategies to products in the NPI Phase • Apply statistical models at scale

  11. Detect Technology & Quality Trends Drove a major shift in test strategy Multi-year view of a design / component margin Trend for 1 electrical test parameter (65 products, 6 years, 100M measurements)

  12. Consistently apply test strategies to a new product Drives the most cost effective test solution up front in the NPI cycle Launch Lean vs. launch and optimize Test strategy dashboard that facilitates choosing the best test optimization strategy for a new product

  13. Apply statistical models / algorithms at scale Determine component variability vs. design margin with higher confidence Reduce non-value-add tests Cpk analysis to determine critical measurement parameters

  14. What can’t we do yet but are working on?

  15. Use big data and analytics to: • Leverage unstructured / semi-structured data • Perform predictive modeling

  16. Leverage Unstructured, Semi-Structured Data Enhance ICT test log

  17. Predictive Modeling Mock up of linear regression attempting to correlate component types to overall product failure rates

  18. Spatial Analytics

  19. Questions?

More Related