John M. Wargo Senior Engineer and Program Manager IBM Supply Chain Engineering October 6, 2016

An Early Warning System for Supply Chain Quality Assurance

©©2016 2012IBM IBMCorporation Corporation

Quality Early Warning System Team  John Wargo

 Paul Zulpa  Jeff Komatsu  Emmanuel Yashchin  Wen Ming Lim

 Reynaldo Corral  Aaron Civil  Tony Spielberg

2

© 2012 IBM Corporation

Agenda

Why a Quality Early Warning System What is QEWS? Examples Summary

Note that the contents/agenda items are written in sentence case Title the page “Table of contents” if the document is meant to be read or is a “leave behind.” Use “Agenda” if the document will be presented formally This page should appear at the beginning of each section, with the highlighted section appearing in blue and bold

Source: If applicable, describe source origin

3

© 2012 IBM Corporation

For quality problems, Time is Money  In any supply chain or manufacturing process, it’s vital to detect quality problems as quickly as possible.

 Even a small delay in detecting a quality problem can have big costs: – – – – –

reworked or scrapped product recall of defective product higher costs in warranty claims loss of customer satisfaction delayed product shipments

Source: If applicable, describe source origin

4

© 2012 IBM Corporation

Quality problems are expensive “The grounding of the Dreamliner fleet has cost Boeing an estimated $600 million, halted deliveries of the aircraft and forced some airlines to lease alternative planes.” (www.reuters.com, 4/27/2013) “The worldwide financial repercussions of Toyota’s recalls could surpass $5 billion over the coming year, primarily because of additional incentive efforts, litigation costs, and marketing campaigns by the struggling auto maker, industry analysts say.” (www.automobile.com, 3/11/2010) “To recap, Nvidia has already taken charges ... totaling over $450 million to cover the costs associated with the warranty, repair, return, and replacement of laptops affected by a "weak die/packaging material set" in certain graphics processing unit (GPU) products.” (www.news.cnet.com, 9/30/ 2010) “What Sony couldn't have foreseen, however, was the massive battery recall that ended up costing them a whopping $429 million and sent their bottom line screaming into the basement.” (www.pcmag.com, 10/26/2006)

5

© 2012 IBM Corporation

The new realities of product development put even the best companies at risk of major quality problems

High volumes of defective product in the field

Relentless cost pressure Higher product volumes Front-loaded product launches Faster product cycles

6

 fewer/overloaded Engineering/Quality staff  use of low-cost components  use of low-cost manufacturers  loss of visibility into manufacturing

 subtle problems are “needles in the haystack” and go undetected  small failure rates translate to large numbers of defective product

 product field population is built “blind”  little or no opportunity to correct mistakes

 less time to qualify new suppliers  less time to qualify new technologies  less time to find design sensitivities  less time to reveal manufacturing problems  less time to get customer feedback

© 2012 IBM Corporation

Agenda

Why a Quality Early Warning System What is QEWS? Examples Summary

Note that the contents/agenda items are written in sentence case Title the page “Table of contents” if the document is meant to be read or is a “leave behind.” Use “Agenda” if the document will be presented formally This page should appear at the beginning of each section, with the highlighted section appearing in blue and bold

Source: If applicable, describe source origin

7

© 2012 IBM Corporation

What is the IBM Quality Early Warning System?  An enterprise-level system which uses unique IBM technology to detect and prioritize quality problems earlier and more definitively than can be done using traditional techniques of statistical process control.  QEWS can be applied at any point where a test, measurement or inspection is made: – In the supply chain • Supplier’s final test • Incoming inspection of raw materials • Incoming inspection of procured components – In manufacturing • At individual production operations • At final product test – In product field performance • Warranty claims

8

© 2012 IBM Corporation

QEWS Analysis Capabilities & Data Types

Module

Typical Data

Typical Applications

Attribute data

• failure rates • yields • sort categories

• Monitoring quality of components procured from suppliers • Monitoring quality of manufacturing operations via inline tests • Monitoring quality of manufacturing operations via final product test

Parametric data

• machine controls • process controls • dimensional measurements • electrical measurements

• Monitoring stability of manufacturing process operations • Detecting shifts in product performance

Reliability data

• warranty claims • stress tests

• Monitoring quality of product in customer use environments • Detecting product wearout

9

© 2012 IBM Corporation

Foundation of the Quality Early Warning System

Automated push of results Legacy Systems Integration Data drill-down tools

IBM, 1984 – 2012

Dashboard Alert Prioritization Setting Thresholds CUSUM 10

E.S. Page, 1954 © 2012 IBM Corporation

Three analysis techniques on the same data

11

© 2012 IBM Corporation

IBM Quality Early Warning System (QEWS) Architecture

Supplier, Manufacturing, Field A

B C

Target optimization algorithm

Strategic data sources: links and data formats

Innovative, very large-scale data cube deployment

Data store and dashboard

QEWS Engine

Prioritization algorithm

Patent pending

12

Threshold setting algorithm

User interface, dashboard and analytics (supplier and OEM) © 2012 IBM Corporation

QEWS Dashboard View History

13

View Trend data

View AVT data

View QEWS data

© 2012 IBM Corporation

View Data Detail

View Comments

Create Ticket

QEWS provides automated, proactive alerts

Dashboard detail End-to-end tracking and alert closure

Push to suppliers

Automatically extract issue part numbers

Alerts proactively sent to end users

Focus parts: pervasive issue, new issue Push to owners

14

© 2012 IBM Corporation

Agenda

Why a Quality Early Warning System What is QEWS? Examples Summary

Note that the contents/agenda items are written in sentence case Title the page “Table of contents” if the document is meant to be read or is a “leave behind.” Use “Agenda” if the document will be presented formally This page should appear at the beginning of each section, with the highlighted section appearing in blue and bold

Source: If applicable, describe source origin

15

© 2012 IBM Corporation

QEWS Demonstration Results: Earlier Detection

This chart shows QEWS analysis results for a set of quality data. The x-axis is aligned in time to the chart below. QEWS alerts when the cumulative evidence crosses above the horizontal threshold line (in black.)

This chart shows SPC analysis results for the same set of data as above. The x-axis is aligned in time to the chart above. SPC alerts when a point falls outside the control limits (at the extreme right-hand side of the chart.)

QEWS alerts 8 weeks earlier than SPC.

16

© 2012 IBM Corporation

QEWS Demonstration Results: Definitive Detection For this set of data, QEWS alerts, then stays in alert mode (above the horizontal black threshold line.) The positive slope of the cumulative evidence line indicates the quality problem is getting worse. From the first alert onward, QEWS presents a clear message that action should be taken.

For the same set of data as above, SPC alerts once, then does not alert again until many points later. Many engineers would dismiss the first alert as an anomaly, and not take action until the second alert.

17

© 2012 IBM Corporation

QEWS Parametric Detection

18

© 2012 IBM Corporation

QEWS Lifecycle Field Warranty Detection

19

© 2012 IBM Corporation

QEWS Lifecycle Field Warranty Detection

20

© 2012 IBM Corporation

Agenda

Why a Quality Early Warning System What is QEWS? Examples Summary

Note that the contents/agenda items are written in sentence case Title the page “Table of contents” if the document is meant to be read or is a “leave behind.” Use “Agenda” if the document will be presented formally This page should appear at the beginning of each section, with the highlighted section appearing in blue and bold

Source: If applicable, describe source origin

21

© 2012 IBM Corporation

QEWS Summary

 Provides a global tool and process (common across the enterprise)  Manages large amounts of quality data automatically, with – earlier alerts – definitive alerts – intelligent prioritization  Empowers engineering staff to focus on the most important problems

 Reduces problem identification time – Faster, preventive problem resolution – Lower total cost of quality

22

© 2012 IBM Corporation

Questions

23

© 2012 IBM Corporation

24

© 2012 IBM Corporation