zero defect and risk mitigation with advanced...

Post on 26-Mar-2018

223 Views

Category:

Documents

5 Downloads

Preview:

Click to see full reader

TRANSCRIPT

Zero Defect and Risk Mitigation with

Advanced Analytics

Joy Gandhi, CQO

Anil Gandhi, Ph.D. President and Chief Data Scientist

Qualicent Analytics, Inc

Agenda

• Qualicent Introduction

• Relevant Trends in the Automotive Industry

• Role of Data and Advanced Analytics

• Technical Goal and the Analytics Process

• Case studies

• Advanced Analytics for Risk Mitigation in the APQP

• Enhanced Manufacturing Anomaly Detection through Analytics

• Summary

Qualicent Introduction

• Services – Advanced Analytics

– Quality Engineering/Failure Diagnostics

– Big Data/IoT Data Integration

• Software – ZeroDefectMiner® software for all industries

– ZeroXMiner for healthcare and IoT

• ABATE Risk Software-Service package

Source: McKinsey

http://www.mckinsey.com/client_service/semi

conductors/latest_thinking

Electronics in Automotive…

…are pervasive

A modern navigation and control panel in a high-end connected car. These type of cars have as many as “50-75 ECUs making them truly distributed computers on

wheels.” (T.Johnson et.al, Univ of Texas Dept of Computer Science and Engineering)

Recall Trend

Significant increase in the recalls

Source: 2015 Automotive Warranty and Recall Report, Stout, Risius and Ross

Note: SRR defines electricals as ignition module and switch, starter assembly, battery, instrument panel, various wiring

EWR Trend - Electricals

Significant increase in the EWRs with injury/death

from electricals

Source: 2015 Automotive Warranty and Recall Report, Stout, Risius and Ross

Confirmed by other Researchers: T. Johnson, et.al. University of Texas Dept of Computer Science and Engineering) and University of Waterloo (Dept of ECE) Paper confirms the clear

rise in the electronic/electrical hazards and risk related notifications in motor vehicles in US, Canada and Europe.

The Solution is in your Data

Evolution of Advanced Analytics

Why is it difficult to achieve Zero Defects?

Business Problem = Reducing Risk

RISK COST

Manufacture

Risk from manufacturing

Risk from bad design

Field Failure

…this presentation

Human Errors

Systemic

Reducing Risk from Manufacturing

RISK COST

Contain

Prevent @Process @ Suppliers

@Manufacturing

Manufacture Field Failure

Marginalities = Units that pass SPC for each and all tests

…but with all tests taken together the unit might be at…

Predictors for large excursions / large effects not difficult to source…BUT

× Biggest field failure losses are from marginal effects and/or intermittent deviations over extended periods

× Marginal effects are difficult to detect with standard methods because of high dimensionality, noise, small # of fails, …

14

OUTLIER 6 s 2 s

6 s 6 s

2 s 2 s __

?

Need advanced methods to detect anomalous parts

o 1000s of components 10,000s of solder points, 100s part SKUs & suppliers

o Each parameter could be within tolerance but combination of parameters may be an outlier

o Lots of available multi-variate combinations which can make the unit an outlier

Source: Mentor Graphics, 2012

Complex devices = Large number of influences / dimensions

Interactions

thickness reflectivity

resistance

capacitance

settling

time

….

Impossible to model on physics

(too many interaction possibilities)

(skewed dataset)

Small number of fails

Tolerances based on individual parameters

Ship marginal product 6 s 2 s

6 s 6 s

2 s 2 s __

?

Yes!

Analytics Process Summary

Traditional: ANOVA, t-test

screen / coarse reduce

Composite distance

cluster analysis

visualization / client

Machine learning model

1. Operating and exclusion

zones for design

2. Anomaly detection

Case Study 1

Field Failure KPI

Composite Distance How

Detect field failures with high class

purity

Result

Who Automotive Semiconductors

VarZ

VarY VarX

Outlier

yes no

Co

mp

osite

Dis

tan

ce

pass fail

pass 6,974 15 6,989

fail 0 2 2

6,974 17 6,991

pass fail

pass 6,981 8 6,989

fail 1 1 2

6,982 9 6,991

@7 @6

Yield Hit = 0.2%

predicted

actu

al

predicted

actu

al

Co

mp

osite

Dis

tan

ce

pass fail

pass 6,974 15 6,989

fail 0 2 2

6,974 17 6,991

@6

Yield Hit = 0.2%

Topmost parameter

Co

mp

osite

Dis

tan

ce

Incumbent Method – Risk Assessment

Project the number of units that will likely fail in the field in the next 10 years

Distance Method – Risk Assessment

“These” units that will likely fail in the field in the next 10 years

?

UCL = Median + x * robust sigma

Accuracy Purity

Composite distance

Top Parameter

Case Study 2

Field Failure KPI

Composite Distance How

Detect almost all field failures with

high class purity

Result

Who Electronic Manufacturing

Co

mp

osite

Dis

tan

ce

To

pm

ost p

ara

me

ter

median + 6*robust s

USL

Five out of seven field failures are detected by Composite Distance…at low cost

Co

mp

osite

Dis

tan

ce

To

pm

ost p

ara

me

ter

pass fail

pass 18,399 5 18,404

fail 2 5 7

18,401 10 18,411

predicted

actu

al

pass fail

pass 18,288 116 18,404

fail 3 4 7

18,291 120 18,411

predicted

actu

al

Composite Distance offers significant improvement over single parameter controls

Pattern Discovery

Deductive Reasoning

Inductive Reasoning

1. Make a hypothesis based on prior knowledge

2. Test the hypothesis

1. Discover patterns, discover hypothesis

2. Check if patterns have material meaning

DISCOVER PATTERNS IMPOSSIBLE TO HYPOTHESIZE

Machine Learning

Traditional Statistics

Field fail, yield, quality,

safety and

effectiveness metrics,

Thickness, resistance

capacitance, time,…

Strategic

Tactical

Model Discovery

Anomaly Detection

y = f(x)

x = x’

INPUTS OUTPUTS

When: Development, Pre-launch, Early production, HVM

Why: Process optimization, Exclusion Zones

How: Exact Models based on machine learning

Class: Supervised

When : HVM

Why: Containment, feedback to suppliers for prevention

How: Iterative Distance methods

Class: Unsupervised

Case Study 3

Large Semiconductor Company Who

Yield KPI

Machine learning algorithms How

Revenue increase by > $ MM/quarter Result

Rule Discovery

Variables M, Q and T individually have no influence on Metric of Interest (MOI)

Data is normalized, scaled and transformed

Variable M Variable Q Variable T

0.0

0.2

0.4

0.6

0.8

Yield = 0 Yield = 1

100 150 200 250 300 700 750 800 850 900 9950 10000 10050 10100

M < 191

Q < 812

T > 10,006

100 150 200 250 300 700 750 800 850 900 9950 10000 10050 10100

0 1

+ +

Variables M, Q and T interactively strongly influence the output

Variable M Variable Q Variable T

Rule Discovery / Machine learning

RESULT:

EXCLUSION ZONE

Case Study 4

PV Solar Company Who

Cell Efficiency KPI

Machine learning algorithms How

Prevent cell efficiency loss by 30% Result

Solar Panel Line Flow

I J K L

Measurement at four sites all passing inspection but low cell efficiency

Algorithms discovered that it’s the ratio that matters

= PATTERN DISCOVERY

Parameters A, B, C, D fully in control and within normal distribution

E F G H

A B C D

Case Study 4

Before Date X

After Date X A

C

Machine learning algorithms discover ratio of A/C as critical parameter (not predicted

by domain experts, but later successfully explained by experts)

EXCLUSION ZONE: Y - low process metric readings (< 24.5) X -low in line measure(< 81) Z (date) > something

Case Study 4: Solar

Machine learning model predicts ~31% reduction in EFF in exclusion zone

Advanced Analytics in the Entire

Product Lifecycle

Proactive Analytics for Risk Mitigation at every APQP Phase

Solution: Analytics in the Product Lifecycle

Advanced

Analytics Product and

Process data

Warranty and

supplier data

Pattern discovery Anomaly detection

Pre-prototype Phases

Concept Product Design

Process Design

Verification

Define requirements

Select materials, suppliers

Identify similar products

Get relevant historical data

Model Discovery

Rules discovery

Adjust process or design to rules for zero defect

Design product, process

FMEA and DFX

Supplier qual data

Anomaly detection on supplier material and process data

CA on material and process

Optimize product and process design

Test device function

Predictive modeling functional and application data; anomaly detection

Optimize product and process design based

Corrective action on anomalies

Data Sources and Outcomes

Analytics DFM, Process and Design CA/CI

Relevant Historical

Warranty/Field Failure Data

Special/Critical Parts/Process

Data

Special/Critical Materials

Supplier Data DFMEA special function/dimensi

ons

Process FMEA special process characteristics

Datasheet Specification

Control

Composite Distance Machine Learning

Outcome Outcome

Techniques

Validation, Safe Launch and HVM

Validation Safe

Launch/SOP High Volume Production

Customer Qualification

Validation/Application Testing

Predictive modeling

Rules discovery

Anomaly detection

Optimize Design

Manufacturing process corrective actions

Optimize Yields

Anomaly detection on pilot

Predictive Model Refinement

Corrective Action on material and process

Optimize product and process design based on predictive model

Ongoing Production Test and Inspection

Anomaly detection on test data

Anomaly Detection on supplier data

Corrective action on anomalies

Corrective Action on maverick/high DPPM lots

The Composite Distance technique has been proven to accurately detect field

failures from manufacturing data.

COMPOSITE DISTANCE CHART

Data involved key OEMs, Tier 1, Tier 2 and Tier 3 suppliers

Worldwide Studies

Composite Distance: Cost Impact

• 7 out of 10 field failures have been detected

• Cost Analysis per Part in a Tier 2 supplier

– Typical electronic board

– 1 year period, 91 failures at Tier 1 and OEM

– Estimated total cost of failure handling =$1.7M

– Cost savings from detection of 70% of failures~$1M

• Impact to reputation and loss of business are not included

Composite Distance Use Cases for SQM

Supplier 1

Server

Data Mirror Data Mirror

Supplier 2

IQC In-process Test

Engineering Mfg

Early Warning Process for Containment

Sample OOC Action Plan

OOC detected

Put product on hold

Risk?

Perform FA

Onsite Eng Dispositions

Eliminate Root Cause

Purge or Recall

Low

High

Medium Stress to fail

Variables of Importance

Process is crucial for full

Issue Resolution

Design Verification Validation HV Production

Pre-proto-type A, B samples C, D Samples Production

• Model Historical Data • Extract operating and

exclusion zones • Improve product and

process design • Anomaly detection on

supplier data

• Model with A,B data • Extract operating and

exclusion zones • Calculate DPPM • Improve product and

process design

• Model with C, D data • Extract operating and

exclusion zones • Outlier Detection for Safe

Launch • Improve process for Safe

Launch

• Ongoing Outlier Detection

• Continuous improvement of Process/product

Prevent Prevent Prevent

Contain

Resolve

Contain

Resolve

Predictive Models Predictive Models

Anomaly Detection (Supplier Data)

Automotive

Sample

Phase

Advanced

Analytics

Goal

Predictive Models

Anomaly Detection

Explanatory Models

Anomaly Detection Rule Discovery

Summary

• Zero defect can be achieved using Advanced Analytics

– Anomaly Detection – unsupervised learning

– Machine Learning – supervised learning

• Contain high probability field failures using composite distance analysis

• Defect reduction and yield improvement can be achieved with predictive models

• Root cause identification with explanatory models

Advanced Analytics can be employed in the entire product life-cycle.

THANK YOU!

BACK-UP

Sample Data Stack for Analytics

Unit # Solder Volume

Reflow Temp

Gas Flow

R12 C48 Shorts Bridging Idd Leakage Func Field

1

0

1

Process Device Defect Inspection Final Test Field

Model/Pattern Discovery

• What are the predictors of DPPM?

Rules Discovery • What are the best operating or process

conditions to achieve low field DPPMs

Anomaly

Detection

• Which parts are highly likely to fail in the field?

Supervised Learning

Supervised Learning

Unsupervised Learning

top related