1
Advanced Infotronics Technologies for
Smart Life Cycle SupportA Closed-Loop Acquisition
Jay Lee
Wisconsin Distinguished and Rockwell Automation Professor andDirector of Center for
Intelligent Maintenance Systems (IMS)
A National Science Foundation (NSF) Industry/University CooperativeResearch Center (I/UCRC)
www.imscenter.net
Simulation-based Acquisition/Advanced Engineering Environment Conference
2
Medium
GPM
Low
GPM
TodayToday
Individual Skills
Internet-Assisted
DigitalTransaction
On Site
Remote
Electronics
e
TomorrowTomorrow
Trends in Maintenance
Ref: IBM
3
Competitive Intelligence
Focus on“Affordability,
Survivability, andSustainability”
Focus on “Smart Service”
&Innovative Life Cycle Support.
Design ProductsManufacturing
andSupply Systems
After Market
ProducibilityPredictability Productivity
Core Intelligence
PollutionPrevention
Performance
Focus on“Velocity”
4Jay Lee
Evolution in Product, Manufacturing, and Quality
ProductFocus
Manufacturing Focus
QualityFocus
1990 2000
Product That
Thinks and Links(information &
computer intelligence)
EnterpriseIntegration
(agility)
Six-SigmaFor
Business Process(enterprise)
1980
FactoryAutomation(flexibility)
SPC, TQM, TPMfor Manufacturing
Process(factory)
Intelligent Mechatronics
(data & controlintelligence)
2010
ProductsThat Learn,
Grow, Reconfigure,and Sustain
(knowledge & distributed Intelligence)
Near-Zero-Downtime &Sustainability
and Asset Optimization(customers)
E-Logistics & E-Manufacturing
(velocity )
IMS Center
5
PredictiveIntelligence toInformate Decision
The Business Model
Execution
DemandInventory Capacity
Inventory Capacity
Execution
Demand
6
Infotronics Technologies intertwine advanced
information and electronics systems
intelligence and enable autonomous business
functions and objectives through the use of
internet and other tether-free technologies (i.e.
wireless, web,…)
Infotronics Technologies ?
IMS Center
7
Current Industry Maintenance Practices
•Reactive >50-60 %
•Preventative >40-50 %
•Predictive <5%
5x
10x
Predictive Preventive Reactive
Cost of Maintenance
“ 60%“ 60%
of all planned maintenanceof all planned maintenance
is unnecessary.”is unnecessary.”
ARC AdvisoryARC Advisory GroupGroup
Fail & FixRoutine maintenance
Fix before Fail
8
Monitoring
Sensors & Events
Product orSystemIn Use
Vision
PredictivePerformance
• Web-enabledmonitoring,Prognostics, anddiagnostics
• Agent for Data Mining
• Logistics IntegrationTools
• Asset Optimization
• Knowledge Mgt.
Acquisition
RobustDesign
Enhanced Six-Sigma Design
Design for ReliabilityAnd Serviceability
Product
Redesign
Web-enabled D2B™ Platform
Degradation Watchdog Agent™
Self Maintenance
RedundancyActivePassive
Communications
• Tether-Free(Bluetooth)
• Internet•TCP/IP
9
– Watchdog Agent™ for Predictive Prognostics
– Web-enabled Smart D2B™ (device to business)Platform and Tools for Data Transformation,Optimization, and Synchronization
– Applied Wireless Technologies
4. Logistics Infotronics Agent (LIA)
IMS Core Technologies
10
Need for Prognostics
Degradation
Process ?
Prognostics
Watchdog
?MODEL
?
Need On-Site Experts
Sensors Monitoring
Signal ProcessingDiagnostics
Action
Assessment
Remote Experts
11
Watchdog Agent
12
Time-Frequency MomentsTime & Frequency Domain
Confidence Value- CV Principal ComponentsMahalanobis Distances
Correlation of multiple signals intoPerformance Prediction- CV
Sensor signals
13
Performance Index Model Based PrognosticApproach
Step3: Degradationtendency prediction
Step 2: Currentperformance calculation
Step 1: Performanceindex modeling
Using parameters fixedafter each maintenance:
Normaldata
Failuredata
)(}1Pr{ 1 xxrv
f=Probability of failure:
Using historydata:
Beginningsetting
Acceptablelevel
)()( 2 xxrr
fC =
Confidence:
Determine model structure ofprobability or confidence curve
With Box-Jenkins method
)()1()( 1 pkykyky p −++−= φφ L)()1()( 1 qkkk q −−−−−+ εθεθε L
On-line parameters ( )
identification, then predict
iφ,iθ
Failure Line
Predict failure probability orconfidence continuously
Currentinputs x
r
}1Pr{ xr
Failureprobability
)(xvC
Confidence
M
14
Mean Time Between Degradation
Trigger pointPredicted degradation
time
CV1
Degradationthreshold
MeanTimeBetweenDegradation
∫∞
⋅0
)( dttPt
time
MTBD =
15
0 50 1 0 0 1 5 0 2 0 0 2 5 0 3 0 0 3 5 0 4 0 00
0 . 5
1
T i m e / d a y s
Prob
abil
ity
of f
ailu
re
0 50 1 0 0 1 5 0 2 0 0 2 5 0 3 0 0 3 5 0 4 0 01
1 . 5
2
2 . 5
3
T i m e / d a y s
Max
spe
ed/m
ps
0 50 1 0 0 1 5 0 2 0 0 2 5 0 3 0 0 3 5 0 4 0 02
2 . 5
3
3 . 5
4
T i m e / d a y s
Cyc
le T
ime/
s
m a i n t e n a n c e
m a i n t e n a n c e
m a i n t e n a n c e
po in t1 po in t2
Fa i lure Line p red ic t ion2 p red ic t ion1
Example
16
System/ Machine
Hybrid Prognostics and Diagnostics
PerformanceStatusWhen
degradationoccur
Product NervousSystems
Look uptables
PerformanceMeasure
Hashing
Degradation (Prognostics)
signal fordegradation
Production
Sensors
whenVirtual Instruments
FusionDiagnostics
+Prognostics
wherewhat
why
Fusion
Diagnostics
Identifyreasons
Internet
Neural NetworkGenetic AlgorithmStatistic ComputingFuzzy Logic…..
17
MACHINE 2
Similarity Matrix for Performance Comparison
.75
--
G
--
HFEDCBA
00.140.750H
-00.29010G
--.670100F
--.11.670.14E
--0.29.60D
--00C
--0B
--A
MA
CH
INE
1
18
Simulation of Throughput Response SurfaceBetween Parameters
0.90.95
10.9
0.95
1
136
138
140
142
144
146
148
150
0.9
0.95
1
0.9
0.95
1
115
120
125
130
135
140
145
150
Sensitivity
Degradation of (X1 & X2)
Sensitivity
19
Watchdog Toolbox
Watchdog ToolboxJoint T-F Method
Applied Statistical MethodsHybrid Techniques
CMAC LearningMethod
20
D2B™ PlatformEmbeddedMonitoring
Prediction & Performance Assessment
Optimization
Synchronization
e-Business,e-Manufacturing,
e-ServiceAutomation
Optimization ofProduction,
Delivery,Maintenance,
Human Resources,
Logistics Systemsetc.
DATA/INFORMATIONDevice/Factory Level
Transformation intoKNOWLEDGE
Real TimeDECISION/
Synchronization
Web
Data/Information Transformation XML
21
Applied Wireless Issues
• Communication Network
– Wireline vs. Wireless
– Performance -Throughput vs. Latency
– Reliability -Control vs. Monitoring
– Cost - ISM Band
• Wireless Technologies
– IEEE 802.11
– Bluetooth (IEEE 802.15)
– Proprietary
Attenuation (dB)
STA
AP d
Building Layout - Sheet Rock Walls
Attenuation (dB)
Building Layout - Cinder Block Walls
Radio Propagation Issue
Prof. Ivan Howitt, EE UWM
22
Tether-Free Embedded Infotronics Agent
HiDRA™ from Rockwell Scientifics
HiDRA™
23
Remote Monitoring and Prognostics ofDefenses Equipment
For short distancetransmission:- 802.11- Bluetooth- etc.
Cellular phones or GPS for longrange transmission
AlarmsPrognostics
Remote diagnosticsService Dispatching
Synchronization
24
Logistics Infotronics Agent (LIA)
Usage, Service & Maintenance
1. Initial product info is written in the product
embedded device
Product delivery
Internet
LogisticsInfotronics Agent (LIA)
Producer’sProduct Life
CycleIntelligence (LCI)
System
5. Design Innovation through life cycle
intelligence
4. Update product usage information in the Life Cycle Monitoring System
3. Wireless Internet connection
between mobile device and
producer’s product life cycle system2. Wireless
bluetooth connection between mobile
device and product embedded device
Tether-FreeTransmission
25
Logistics Infotronics Agent
(LIA)
26
Final ProductAssembly and
Testing
Supplier 1e-Shops
Enterprise Database
Mgt System (EDMS)(design, manufacturing, suppliers, testing, etc)
Supplier 2e-Shops
Supplier Ne-Shops Sourcing
Life Cycle Service &Maintenance (LCSM)
Supplier Evaluation& Selection
….
Dec
isio
n
e-Procurement
Manufacturing Enterprise Execution System (MEES)
Product, Process, and EnterpriseDesign (PPED)
CustomerNeeds
Today’s Acquisition and Logistics Systems
27
Final ProductAssembly and
Testing
Supplier 1e-Shops
Logistics InfotronicsAgent (LIA)
Data Accessibility&
Monitoring
EnterpriseDatabase
Mgt System (EDMS)(design, manufacturing,suppliers, testing, etc)
Product, Process, and EnterpriseDesign (PPED)
Supplier 2e-Shops
Supplier ne-Shops Sourcing
Life Cycle Service &Maintenance (LCSM)
Supplier Evaluation& Selection
….
Dec
isio
n
e-Procurement
Manufacturing Enterprise Execution System (MEES)
Tether-free Interface
Prediction, Validation & Optimization
CustomerNeeds
Design for Six-Sigma(DFSS)
SupplierQuality
Tether-free Interface
28
• National Instruments
• ITRI• PMC
• Kone Elevator• API• Genex Technologies
• ATOP
• Dr. Machine.Com
• Hitachi Seikei
• Velicon
• Eagle Technologies
• Industrial Objects
• InterNext Group
• Citation Custom Products
Current Members and Sponsors
• Rockwell Automation• Ford Motor• GM• Johnson Controls• Eaton• Harley-Davidson
• Xerox• Wisconsin Electric Power• United Technologies• Siebel Systems• GE Medical Systems• Toshiba (Japan)• A.O. Smith• Intel• U.S. Postal Services
• DaimlerChrysler
• Lantronix
• Micro Mobio Wireless
• Cognex
• Boeing
• DoD DMSO
• Motorola
• ABB
• Caterpillar
• SEMATECH
29
Testbeds andProjects
CompanyMembers
Knowled
ge, B
est P
racti
ces,
Tools
, and
Stud
ents
IMS Methodologies and Enabling Technologies
Testbeds andProjects
Value through Leveraged Partnerships
Membership and Partnerships
30
Why IMS?
IMS Methodologies
Fail and Fix orFly and Fix Predict and
Prevent
FaF PaP
IMSTestbeds
(e-Maintenance/e-Manufacturing/
e-LogisticsIntegration)
Partnerships
Risk Reduction, Technology Readiness,
andNew Breed of
Engineers and Workforce
31
For More Informationwww.imscenter.net
Thank You