march 5 - 8, 2017 hilton phoenix / mesa hotel mesa ......bits 2017 burn-in & test strategies...

BiTS 2017

March 5-8, 2017 Burn-in & Test Strategies Workshop www.bitsworkshop.org

Keynote

Archive - Keynote

March 5 - 8, 2017

Hilton Phoenix / Mesa Hotel

Mesa, Arizona

© 2017 BiTS Workshop – Image: tonda / iStock

BiTS 2017


Copyright Notice

The presentation(s)/poster(s) in this publication comprise the Proceedings of the 2017 BiTS Workshop. The content reflects the opinion of the authors and their respective companies. They are reproduced here as they were presented at the 2017 BiTS Workshop. This version of the presentation or poster may differ from the version that was distributed in hardcopy & softcopy form at the 2017 BiTS Workshop. The inclusion of the presentations/posters in this publication does not constitute an endorsement by BiTS Workshop or the workshop’s sponsors. There is NO copyright protection claimed on the presentation/poster content by BiTS Workshop. However, each presentation/poster is the work of the authors and their respective companies: as such, it is strongly encouraged that any use reflect proper acknowledgement to the appropriate source. Any questions regarding the use of any materials presented should be directed to the author(s) or their companies. The BiTS logo and ‘Burn-in & Test Strategies Workshop’ are trademarks of BiTS Workshop. All rights reserved.

BiTS 2017


Keynote

BiTS Workshop 2017 Schedule

Frontiers Day Monday March 7 - 9:00 am

Identifying the path to success for Industry 4.0

Thomas Sonderman

Rudolph Technologies

Sponsored by

Keynote

Address

BiTS 2017


Keynote

Identifying the Path to Success for Industry 4.0

Thomas Sonderman

VP & GM, Integrated Solutions Group

BiTS 2017


Keynote

Market Driven Challenges

5

SOURCE: NCTA

BiTS 2017


Keynote

What is Industry 4.0? Wikipedia…

“… cyber-physical systems monitor physical processes, create a virtual copy of the physical world and make decentralized decisions.

Over the Internet of Things, cyber-physical systems communicate and cooperate with each other and with humans in real time…”

6

BiTS 2017


Keynote

Components of Industry 4.0 Attributes of all these connected “things”

• Discoverable

• Autonomous

• Model-based

• Communicative

• Self-monitoring

• Secure

• Standards-based

Imagine the collaborative behavior that could emerge !

7

BiTS 2017


Keynote

Solutions MUST Enable Industry 4.0 The Drive Towards Increased Productivity The evolution toward smart manufacturing is enabled by the confluence of

Connectivity

Computing Power

Intelligent Software

Ability to Manage MASSIVE DATA

Electronics Suppliers Who Act Decisively will Benefit

by Creating Added Value for their Customers

8

BiTS 2017


Keynote

Key Challenges in Bringing ICs to Market

Data Democracy • Multiple design groups

• Challenging wafer fab,

Assembly/Test and system

integration

• Complex entity name changes

Supply Chain Management

Design for Profitability • Product concept/circuit design

• Reticle set design

• Product design debug &

optimization

• Test program debug

Product Management

Ramp and Availability • Multiple data consumers

• Design

• Foundry management

• Customers

• Suppliers

Manufacturing

Real-Time Web Based

Reporting and Drill Down Layout Aware Analysis Fab-to-Field / Field-to-Fab

Integrated Solutions are Required

9

BiTS 2017


Keynote

Understanding Supply Chain Yield Dynamics

Requires Complex Enterprise Level Supply Chain Optimization:

• Big Data

• Digital Threads

• Package to Tool Correlations

• Real-time Dashboards

• Best in Class Yield Analytics

Component Matching

Package Interconnection Delays

Device Parametric Variation

Performance Failures

Structural Failures

Device Contamination

Process Tool Failure

Bump Coplanarity

Solutions Must Remove Risk from

the Electronics Supply Chain

10

BiTS 2017


Keynote

IC Fabrication Information Flow

11

NitrideFurnace

DepositionSTI Masking

ShallowTrench Etch

Metrology Metrology Metrology

TEOS FurnaceDeposition

TrenchPlanarization

Masking

TrenchPlanarization

EtchMetrology Metrology Metrology

STI Polish Nitride StripMetrology Metrology

DepController

CDController

DepthController

Dep Controller(Development)

OverlayController

Etch LossController

ThicknessController

Oxide LossController

Dep Time andTemperatures

Dep Time andTemperatures

ExposureEtch Time

(By Chamber)

EtchTime

PolishTime

BathTimes

NitThk

NitThk

NitThk

NitRef CD

OxideThk

OverlayMeasurements

OverlaySettings

OxideThk

OxideThk

OxideThk

OxideThk

TrenchDepth

TrenchDepth

Flow of Data

Flow of WIP

OxideThk

EndpointTimes

Process Control

and Metrology

Design statistics

& care-abouts

Process Statistics

and capability

Knowledge

Aware Design

Data

Data

Process/Design

Sensitivities

Design-based

Yield Diagnostics

Adaptive Test

Data

Data

BiTS 2017


Keynote

INTEGRATED SOLUTIONS

METROLOGY PROBE/TEST PROCESS

Product State Wafer State Tool state

4

Converting Data into Actionable Intelligence

BiTS 2017


Keynote

Big Data Defined

13

BiTS 2017


Keynote

Why Can’t Decisions Be Made Quicker?

14

BiTS 2017


Keynote

New Paradigms are Required

15

BiTS 2017


Keynote

Electronics Supply Chain Optimization

16

Trace Data

Tool Level

Fab Level

Fab 1 Adv. Package Multi-chip

Product

Centralized

dB’s

Chamber

Process Area

Fab 2

Enterprise YMS

Sophisticated Yield Management System

Supply Chain Traceability (Digital Threading)

Dashboards for Complete Visibility from Technician to CEO

Bi-Directional Communication

Enterprise FDC

Fault Detection with Wafer/Die Level Correlations

Analytics Across the Value Chain

Deep into

the process

Big Data

BiTS 2017


Keynote

Design /IP Optimization Process Optimization Test Optimization

Data Analysis and Correlation

Data Generation and Warehousing

Test Chip

Data

Design

Statistics

Defect

Data

Test and

Diagnostics

Bitmap

Data

Process

Metrology

Equipment

Data

OPC

Data

Productivity

Data

System

Test Data

DPPM

Integrated Yield Improvement Infrastructure

Leakage Perf/Power

Yield/Cost Cycle Time

Control

Metrology

Equipment

Cycle time

DFT

Repair Adaptive test

Burn-in

17

BiTS 2017


Keynote

DATA STREAMS

Intelligent Analysis Engines

18

DATA WAREHOUSE

Defect data Metrology

data

MES &

Equip data

Sort Yield,

Parametric

Design

info

Parametric

Test data

E-Test Tools Defect Tools Metrology

Tools

Control

Tools

Process

Tools

ADVANCED ANALYTICS ENGINE

Control

Real-time

Data Collection

On demand

Analytics

Wafer-centric Tool-centric

Thread

Synchronization

Engines

Deep

Learning

Algorithms

Adaptive

Knowledge

Library

Data Modelling

BiTS 2017


Keynote

Advanced Predictive Analytics PIES Algorithms for Machine Learning

• Probability Bayesian Predictor

• Information Adaptive Decision Trees

• Error Partial Least Squares

• Similarity k-nearest Neighbor

19

BiTS 2017


Keynote

Predictive Analytics for Industry 4.0 Comprehensive Algorithm Suite

20

PCA

Unsupervised Visual

PLS

Supervised Visual

Adaptive Decision

Trees

Supervised Prioritization

MLR

Supervised Quantification

ANOVA

Univariate Validation

Cluster

Un-Supervised Prioritization

Dynamic Time

Warping

UVA / MVA Pre-filter

Multivariate Analysis

Which lots behave differently?

NPI Modeling Algorithms

Which parameters are critical?

Predictive Analysis

What is the predicted yield?

TYPES OF MACHINE LEARNING

BiTS 2017


Keynote

PLS (Partial Least Squares Regression)

Observed variables Predicted variables

• Multi-variate prediction modeling

• Linear Regression: Y ~ X

• Project observed (X) variables and

predicted (Y) variables to a new space

Correlation Chart & Data in new space

Prediction Result

21

BiTS 2017


Keynote

Parameteric Interpolation Thin Plate Spline Regression (TPS)

Measured samples

Predicted values

22

BiTS 2017


Keynote

Decision Tree Mathematics

What is a decision tree?

• Sequential statistical test

• Identify primary cause/effect

• Continuous Variables

• Liner regression or curve fitting

• Categorical Variables

• ANOVA

• Time Based data

• Split data based on primary effect

• Repeat to create tree with 2N branches

• More than one valid predictor

• Show best answer first

• Presentation-ready graphics

23

BiTS 2017


Keynote

Intelligent Digital Threading™

24

FAB 1

B FAB 2

AP

PERFORMANCE

YIELD

ACTIONABLE DATA, NOW! ™

BiTS 2017


Keynote

ACTIONABLE DATA, NOW! ™ METROLOGY

25

PROCESS

STEP

Yield Spec

Target Limit

Recipe

Runtime

Sensor

Eq

History

T.F.

3D

THREAD SYNC ENGINE ™

Spec

Metrology

Images

With Die

Die

Wafer

Lot

Batch

F.T

X-Ray

PRODUCT

EQUIP

FAB 2 ANALYTICS

YMS

Viz

VM YM

SPR

FDC

Defect

BiTS 2017


Keynote

Digital Threading - Multichip Module Helping Automotive Customer Achieve “Zero Defect” Goals

Cap Wafer

Lot: E1234.3

Shipment to Customer

Assembly External

CMOS ASIC House

Test Center

Assembly

Assembly

Final Parts after Assembly (SMB)

Lot: U5678.15

Sensor after waferbonding Lot: E1234.3

ASIC Wafer Lot: U5678.2

WaferBonding

Sensor Wafer

Lot: L58280001.4

Traceability throughout the supply chain driving yield improvement

26

BiTS 2017


Keynote

27

Front End Final

Test Sort

• A long cycle time process – Die bank

• After die bank the lot is split into many lots

• Prior to die bank there are 4 lots moving through the fab. These are

the 4 lots in this multi-chip module

• Wafer fabrication of all

module components

• Traces parent lots and

all children through:

• Sort

• Assembly

• Final test

ASIC Sensor1

Sensor2 Cap

4-Chip Module

Digital Threading – Multi-chip Module

BiTS 2017


Keynote

Post Digital Threading, you can analyze all

components’ data collected:

• Inline parametric

• WET or E-test data

• Sort Bin/Parametric

• Final Bin/Parametric

28

Fail Bin 25 has the

highest ratio at

Final Test. The

corresponding

parametric

parameter is Isb_H,

which was tested at

Sort as well.

MCM – Analysis Of All Manufacturing Data

BiTS 2017


Keynote

Genealogy associates sort lot number with final test lot number

• Combined dataset shows parametric shift due to packaging

Plot the sort parametric value and the final test

patrametric value

• There is a paramteric shift during packaging

• The yield on this multi-chip module is very

sensitive to paramertric variation

FT_2.0_Isb_H

Sort_Isb_H_1

Cum Probability Plot – Parameter @ FT and Sort

29

BiTS 2017


Keynote

Advanced RF Device Digital Threading – Duplexer Example

30

BiTS 2017


Keynote

Advanced RF Device Digital Threading – Duplexer Example

31

BiTS 2017


Keynote

Intelligent Visualization

Yield by week

& location

Yield by

product family

Yield by

layer ID

Yield by

product ID Yield by AOI

tool ID

Yield by

lot ID Yield by class method

& defect class

Filters & display

settings

32

BiTS 2017


Keynote

Industry 4.0 Analytics Platform

33

BiTS 2017


Keynote

Summary

34

• The Goal is Actionable Intelligence

• New Analytical Paradigms will be Required

• Requirements:

• The Right Data Actionable, Secure, Relevant and High Integrity

• At the Right Time Current, Accurate and Transparent

• In the Right Format Visible, Predictive and Prescriptive

BiTS 2017


Keynote

march 5 - 8, 2017 hilton phoenix / mesa hotel mesa ......bits 2017 burn-in & test strategies...

Documents