Intelligent Maintenance System

37
1 Advanced Prognostics for Smart Systems Methodology, Tools, and Applications NSF I/UCRC since 2000 Jay Lee Ohio Eminent Scholar and L.W. Scott Alter Chair Professor Univ. of Cincinnati [email protected] & Director NSF Industry/University Cooperative Research Center on Intelligent Maintenance Systems (IMS) www.imscenter.net Outline Business and Productivity Models Prognostics Design: Methodology and Tools Design of Prognostics Platform for Various Applications Design of Smart Service Systems

description

Intelligent Maintenance System

Transcript of Intelligent Maintenance System

Page 1: Intelligent Maintenance System

1

Advanced Prognostics for Smart Systems Methodology, Tools, and Applications

NSF I/UCRC since 2000

Jay Lee

Ohio Eminent Scholar and L.W. Scott Alter Chair ProfessorUniv. of Cincinnati

[email protected]&

Director NSF Industry/University Cooperative Research Center on

Intelligent Maintenance Systems (IMS)www.imscenter.net

Outline

• Business and Productivity Models• Prognostics Design: Methodology and Tools• Design of Prognostics Platform for Various

Applications• Design of Smart Service Systems

Page 2: Intelligent Maintenance System

2

Outline

• Business and Productivity Models• Prognostics Design: Methodology and Tools• Design of Prognostics Platform for Various

Applications• Design of Smart Service Systems

IMS Vision and Mission Statement

To enable products and systems to To enable products and systems to achieve and sustainachieve and sustain nearnear zerozeroachieve and sustain achieve and sustain nearnear--zero zero breakdown performancebreakdown performance, and ultimately , and ultimately transform machine condition data to transform machine condition data to useful information for improved useful information for improved productivity and asset utilization.productivity and asset utilization.

Page 3: Intelligent Maintenance System

3

The IMS Consortium

Common Unmet Needs in Future Maintenance and Service

Intelligent Monitoring, Predict and Prevent for

Guaranteed SustainabilitySelf-Assessment

Autonomous InformationFlow from Field (user)To Service/BusinessSystems

OperationsIntelligence

Only Handle Information Once (OHIO)

Predict, Prioritize, Optimize, and Plan Maintenance SchedulingFor Near-Zero Downtime and

All-Time Readiness

Page 4: Intelligent Maintenance System

4

Outline

• Business and Productivity Models• Prognostics Design: Methodology and Tools• Design of Prognostics Platform for Various

Applications• Design of Smart Service Systems

Service Design using “System-of-System” Approach

Design for Reliability and Serviceability

Health MonitoringSensors & Embedded

Intelligence

Product orSystem

In UseJust-in-Time

Service

Near “0”Downtime

Loop

Design

Closed-LoopLife CycleDesign

E h d

ProductCenter

ProductRedesign

Self-Maintenance•Redundancy•Active•Passive

Communications •Tether -Free(Bluetooth)

• Internet•TCP/IP

IntelligenceIn Use Service

SmartDesign Condition-based

Maintenance(CBM)

• Web-enabled Monitoring & Prognostics

• Decision Support Tools for Optimized Planning &Scheduling

• Responsive Services

Degradation Assessment(Feature Monitoring)

EnhancedSix-SigmaDesign

•TCP/IP p

• Asset Optimization

Health Information

Page 5: Intelligent Maintenance System

5

Principal ComponentsPerformance Feature

System Instrumentation ApproachSystem Instrumentation Approach

7

7.5

Prediction of Machine Performance Life

Normal Behavior

Most Recent

Behaviorr

Health Feature Radar Chart

1 Bearing

200 210 220 230 240 250 260 270 280

3

3.5

4

4.5

5

5.5

6

6.5

Cycle Number

Prin

cipa

l Com

pone

nt

Actual DataPredictionActual DataPrediction Uncertainty

00.20.40.60.8

1 Bearing

Tools for Autonomic Computing

User input

Page 6: Intelligent Maintenance System

6

ture

f2

Example of two dimensions spaceFailure

Mode #1

Failure Mode #2

Safe operation

Statistical Pattern Recognition• What?

Analytical comparisons of feature distributions

• When?

0Feature f1

Feat Mode #2

Cluster distribution

CV C t

Cluster displacement

When?Features are (approximately) GaussianStable operation without too many impacts

• Possible application:

Time

1

0.5

0

CV Current time

Evaluation the overlap area yields the CV

ppRepeatable processes and operations

Methodology

Smart• Assess Health Degradation• Predict Performance Trends

Streamline• Sort, Filtering, Prioritize Data• Reduce Sensor Data Sets & PCA• Correlate and Digest Relevant Data

Smart Processing

Predict Performance Trends• Diagnose Potential Failure (Prognostics)

Synchronize

• Embedded Agent (hierarchical and distributed)• Only Handle Information Once• Tether-free Communication• Decision Support Tools

Standardize• Systematic Prognostics Selection• Reconfigurable Hardware Platform Toolbox

M i t I f ti St d di ti• Maintenance Information Standardization

Sustain• Embedded Self-Learning & Knowledge Mgt. • Closed-Loop Life Cycle Design• User-Friendly Prognostics Deployment

12

Page 7: Intelligent Maintenance System

7

Data Streamlining

Streamline

Expert k l d

MaintenanceDatabase

No Maintenance

Record

Data Cleaning, Fil i d S iSmart

Processing

Synchronize

St d diData Collection

Expert Knowledge

knowledge Filtering and Sorting

Identify Critical Components

Standardize

Sustain

Data Cleaning, Dimension Reduction

Determine a Clean and Reduced Data Set

IMS Methodology: 5S Approach

• Assess Health Degradation

Streamline• Sort, Filtering, Prioritize Data• Reduce Sensor Data Sets & PCA• Correlate and Digest Relevant Data

Smart Processing

g• Predict Performance Trends• Diagnose Potential Failure (Prognostics)

Synchronize

• Embedded Agent (hierarchical system)• Tether-free Communication• Only Handle Information Once (OHIO)• Decision Support Tools

• Systematic Prognostics ImplementationStandardize

y g p• Reconfigurable Hardware and Software

Platform• Maintenance Information Standardization

Sustain• Embedded Knowledge Mgt. for Self-Learning• Closed-Loop Product Life Cycle Design• User-Friendly Prognostics Deployment

Page 8: Intelligent Maintenance System

8

Watchdog Agent ToolboxSignal Processing & Feature Extraction

Time-frequency / Time-frequency

Performance Assessment

Logistic Regression

Performance Prediction

M t h M t i

ARMA Prediction

moments

Wavelets/Wavelet packet energies / Wavelet moments

Autoregressive modeling / AR model roots

Statistical Pattern Recognition

Feature Map Pattern Matching

CMAC Network Pattern Match

Match Matrix

Elman Recurrent Neural Network

Fuzzy Logic

Health Diagnosis

Expert extracted features

?

Time

1

0.5

0

CVCV

Pattern MatchSupport Vector

MachineHidden Markov

Model

Bayesian Belief Network

Fast Fourier Transform

Information Delivery

Results of Smart Prognostics Tools for Asset Health Information

Confidence Value Health Map for Health Radar Chart

00.20.40.60.8

1

Risk Radar Chart for performance degradation assessment(0-1)

ppotential issues and pattern classification

for multiple components degradation monitoring

toPrioritize Maintenance Decision

Page 9: Intelligent Maintenance System

9

Information Delivery

Results of Smart PrognosticsTools for Asset Health Information

Confidence Value Health Map for Health Radar Chart

00.20.40.60.8

1

Risk Radar Chart for performance degradation assessment(0-1)

ppotential issues and pattern classification

for multiple components degradation monitoring

toPrioritize Maintenance Decision

Field Service Application: KONE Testbed

Equipment Elevator Door

(Controller)

Remote Monitoring

E d

OPC Server

WAGO I/O

iCONICS Genesis32(OPC client)

IIS web server

Web Browser

Bridge

BusinessApplications

Encoder

Watchdog Database

iCONICS GenBroker

(COM over tcp/ip)

Bridge

D2B™

Page 10: Intelligent Maintenance System

10

A Case Study: Elevator Door A Case Study: Elevator Door TestbedTestbed

Performance Indicators

Normal Operation

CV = 97.6%

Normal Operation

CV=97.6%

• Cycle time of the elevator door

• Maximal speed of the elevator dooe

Indicators

Increased Friction

CV=70.3%

Information Delivery

Results of Smart Prognostics Tools for Asset Health Information

Confidence Value Health Map for Health Radar Chart

00.20.40.60.8

1

Risk Radar Chart for performance degradation assessment(0-1)

ppotential issues and pattern classification

for multiple components degradation monitoring

toPrioritize Maintenance Decision

Page 11: Intelligent Maintenance System

11

ESMD Chillers FCFT Predictive Maintenance for

Hong Kong International Airport Control TowerHong Kong International Airport Control Tower (May 2007)

Project Background

Chiller at Control Tower of H.K. Int. Airport

Page 12: Intelligent Maintenance System

12

Hardware Setup

Chiller with 6 AccelerometersDetails:IMI 623C01, high-freq., ceramic shear ICP accel, 100mV/g, 0.8 to 15k Hz, top exit, 2-pin conn.From: S&V Samford Instruments Ltd.

NI PXI Data Logging SystemDetails:- PXI-4472, 24 Bit, 102.4 kS/s, 8 Inputs, 110 dB Dynamic Range- 6 SMB to BNC connectors, SMB-100, SMB to BNC Female- Communicate with Data Logging Server through NI MXI-4 technology

Johnson ControlsOPC Server

35m BNC Cable x 6

3m copper cable3m LAN cable

Data logging ServerDetails:- Dell Dimension(TM) E520- Data Logging System

3m copper cable

Raw Data

Vibration signals OPC (OLE for Process Control) Data

Page 13: Intelligent Maintenance System

13

Fixed Cycle Feature Test (FCFT)

• How to convert raw data to information?

T2 T3information?

• Why FCFT?• What is FCFT?

T1

T – transition of the load change

FCFT Radar Chart

EMSD vibration signals on 2007-05-21 at 16-03-54 ----------- Normal

Data ---- Information

Page 14: Intelligent Maintenance System

14

FCFT Radar Chart

EMSD vibration signals on 2007-05-22 at 10-57-04 ----------- Normal

FCFT Radar Chart

Some mechanical faults happened at that time.

EMSD vibration signals on 2007-05-14 at 11-22-18 ----------- Abnormal

Page 15: Intelligent Maintenance System

15

Information Delivery

Results of Smart Prognostics Tools for Asset Health Information

Confidence Value Health Map for Health Radar Chart

00.20.40.60.8

1

Risk Radar Chart for performance degradation assessment(0-1)

ppotential issues and pattern classification

for multiple components degradation monitoring

toPrioritize Maintenance Decision

KomatsuKomatsu

Earth Moving Equipment Field Services

(Komatsu Engineer on UC Campus)

Page 16: Intelligent Maintenance System

16

Komatsu’s Remote Maintenance• Komatsu Ltd. has developed the necessary

infrastructure for collecting data from individual construction machines, from the

– Failure History data• Failure code, Occurrence date & time,

Repaired date & time– Trend Analysis Data

D t f ( / / i

Help deskplace of operation via the user

• Data from sensors (as max/ ave/ min every 20 operating hours)

– Others data• Load meter, Running time of engine,

etc.

Mode m Bank

Mode m Bank

Mode m Bank

• Failure prediction & diagnosis failures– to reduce down time, maintenance & support

costs

Focus on Smart Asset Business

• by detecting incipient degradations of all components• by identifying and isolating failure critical components

• Degradation monitoring of engine and transmission systems– to estimate lifetime with higher accuracy

t t d l f il– to prevent or delay failures– to expand components overall life

• using appropriate and key data• using features for key performance index

Page 17: Intelligent Maintenance System

17

Komatsu Approach

– to increase quality of data in Komatsu database

– to enable more advanced predictive maintenance decision making base on analytical reliability models

– to achieve intelligent maintenance excellence

Sensors

IMS WatchdogTM KOMATSU On -

Board Computer– and to implement knowledge based decision making approach

Board Computer

Monitoring & Aid View to DecisionEngine DT09XXN Replacement, High priority

Engine DT09XYH Replacement, High priority

Engine DT43XZA Maintenance requiredPlan replacement in 1500 h

Engine DT77YVP Good, 5000 h remaining

Engine

Comp.A,C

Comp.D

No change

IMS Embedded Smart Software Tools

ENGINEERING KNOWLEDGE AND DATAACQUISITION FROM THE SYSTEM

Komatsu WebCARE

Customer Appliance

Komatsu Engineers

Signal Processing&

Feature Extraction

Health Assessment&

Prediction

Bayesian BeliefNetwork (BBN)

C

Autoregressive Modeling

Match Matrix & ARMA Modeling

Group Methods Data Handling Based Feature Extraction

B

Fuzzy Logic (FL) Based Decision Making

D E

Data Preparing (Filtering Outlier estimation)

A

Human MachineInterface

AUTOMATIC AID TO DIAGNOSIS DECISION MAKING

AUTOMATIC AID TOPROGNOSIS DECISION MAKING

Remaining Lifetime Prediction Comparison and Predictive Maintenance Planning

F

Page 18: Intelligent Maintenance System

18

Failure Mode Effects & Risk Assessment Tools

Decision Support Tools for Smart Service Recommendation

FL or Reliabilitybased Prognostics

WearCrackingCable disc.Sensor failsInjectorWear M Change 0.9Derailment Fall of the Shaft 1

CylinderInlet Valve RInlet Valve LInjector…

4 0.6 H Change engine

S Prognostic

Machines

Effect on the system

Machine B

Engine AXMachine A

CCauses

2FM 1

FM 2

Fall of the cabin

Transmit. BY

Component D PSystem ref. Engine ref. Failure Mode

CrackingSensor fails

D: Detectability C: CostP: Probability S: Severity

Bearingof the cabin cabin GearsBearing

BBN output Decision Aid

D

0 40.60.8

1

Autonomic Service Logistics

Valve

Bearing Logistic Information

•Engine/Asset Location

00.20.4Valve

Cylinder

•Components Availability

•Review Previous Maintenance Actions

•Recommend and Assign Right Technician

•Estimate Maintenance•Estimate Maintenance cost

Page 19: Intelligent Maintenance System

19

Information Delivery

Results of Smart Prognostics Tools for Asset Health Information

Confidence Value Health Map for Health Radar Chart

00.20.40.60.8

1

Risk Radar Chart for performance degradation assessment(0-1)

ppotential issues and pattern classification

for multiple components degradation monitoring

toPrioritize Maintenance Decision

Bearing Fault BackgroundNormal Roller Inner-race Outer-race

Outer + inner Outer-race Outer + inner Outer race +roller + roller + roller

Drawing courtesy of Wikipedia

Normal condition Outer-race failure Roller failure Inner-race failure

Page 20: Intelligent Maintenance System

20

Feature Extraction

FFT (Fast Fourier Transform)

Overall acceleration, velocity, or displacement, amplitude peaks at1X RPM, 2X RPM, 3X RPM, BPFO, BPFI, FTF and BSFHarmonics and sidebands, RMS and kurtosis

Health Assessment - Multi-fault Data

Inner- race

Roller

Outer- race

Normal Condition

Outer Race + Roller

Outer + Inner + Roller

Outer + Inner Race

Inner- race + Roller

SOM accomplishes two things: reduces dimensions and displays similarities.

Page 21: Intelligent Maintenance System

21

IMS Methodology: 5S Approach

• Assess Health Degradation

Streamline• Sort, Filtering, Prioritize Data• Reduce Sensor Data Sets & PCA• Correlate and Digest Relevant Data

Smart Processing

g• Predict Performance Trends• Diagnose Potential Failure (Prognostics)

Synchronize

• Embedded Agent (hierarchical system)• Tether-free Communication• Only Handle Information Once (OHIO)• Decision Support Tools

• Systematic Prognostics ImplementationStandardize

y g p• Reconfigurable Hardware and Software

Platform• Maintenance Information Standardization

Sustain• Embedded Knowledge Mgt. for Self-Learning• Closed-Loop Product Life Cycle Design• User-Friendly Prognostics Deployment

Synchronization System Structure

InternetHealth InformationData flow

HealthInformation

Display

ATC StatusDisplay

Discrete EventDisplay

User

Interf

Embedded side Remote side

......

Error LogDatabase

AlgorithmModification

Data Format

ace

Industrial FieldRaw data

Page 22: Intelligent Maintenance System

22

IMS Methodology: 5S Approach

• Assess Health Degradation

Streamline• Sort, Filtering, Prioritize Data• Reduce Sensor Data Sets & PCA• Correlate and Digest Relevant Data

Smart Processing

g• Predict Performance Trends• Diagnose Potential Failure (Prognostics)

Synchronize

• Embedded Agent (hierarchical system)• Tether-free Communication• Only Handle Information Once (OHIO)• Decision Support Tools

• Systematic Prognostics ImplementationStandardize

y g p• Reconfigurable Hardware and Software

Platform• Maintenance Information Standardization

Sustain• Embedded Knowledge Mgt. for Self-Learning• Closed-Loop Product Life Cycle Design• User-Friendly Prognostics Deployment

Watchdog AgentWatchdog Agent®® Software PlatformSoftware Platform

Executive Agent

Task Scheduler ISO 13374 /Engine

AlgorithmToolbox

Algorithm Selector

Health Diagnosis

PerformanceAssessment

PerformancePrediction

Feature Selection

Processing TaskISO 13374 /OSA-CBM

44

CommunicationsToolbox

Database

Agent External Devices

Services

Page 23: Intelligent Maintenance System

23

D2B™ Machine Level Connectivity Sensor Fusion

Feature Extraction

Reconfigurable Prognostics Platform

Performance AssessmentAutonomous Learning

Prediction and PrognosticsDecision Support Tools

D2B™ Business Level Connectivity

IMS Prognostics Platform

IMS Methodology: 5S Approach

• Assess Health Degradation

Streamline• Sort, Filtering, Prioritize Data• Reduce Sensor Data Sets & PCA• Correlate and Digest Relevant Data

Smart Processing

g• Predict Performance Trends• Diagnose Potential Failure (Prognostics)

Synchronize

• Embedded Agent (hierarchical system)• Tether-free Communication• Only Handle Information Once (OHIO)• Decision Support Tools

• Systematic Prognostics ImplementationStandardize

y g p• Reconfigurable Hardware and Software

Platform• Maintenance Information Standardization

Sustain• Embedded Knowledge Mgt. for Self-Learning• Closed-Loop Product Life Cycle Design• User-Friendly Prognostics Deployment

Page 24: Intelligent Maintenance System

24

Next Step Close the Loop

Design for Reliability and Serviceability

Health MonitoringSensors & Embedded

Intelligence

Product orSystem

In UseJust-in-Time

Service

Loop

Design

Closed-LoopLife CycleDesign

E h d

ProductCenter

ProductRedesign

Self-Maintenance•Redundancy•Active•Passive

Communications •Tether -Free(Bluetooth)

• Internet•TCP/IP

IntelligenceIn Use Service

SmartDesign Condition-based

Maintenance(CBM)

• Web-enabled Monitoring & Prognostics

• Decision Support Tools for Maintenance Schedule

• Synchronization Service

Watchdog Agent™Degradation Assessment(Feature Monitoring)

EnhancedSix-SigmaDesign

•TCP/IP

Watchdog Agent is Trademarks of IMS Center

• Asset Optimization

Health Information

Int. IMS Project PROMISE – (Product Embedded Information System for Service and EOL)

1. Initial product info is written in the product embedded device

Product delivery Producer’s

KPDM

Internet

Embeddeddevice 3. Wireless Internet

connection between mobile device and producer’s KPDM

3. Wireless bluetooth connection between mobile device and

product embedded device

Bluetooth ?

Service &maintenance EOL

2. A certified agent for service or EOL operations4. Update information in

the embedded device and in the producer’s KPDM

Page 25: Intelligent Maintenance System

25

Outline

• Business and Productivity Models• Prognostics Design: Methodology and Tools• Design of Prognostics Platform for Various

Applications• Design of Smart Service Systems

Prognostics Design ad Development

Smart

Streamline

Smart Processing

Synchronize

Standardize

50

Sustain

Page 26: Intelligent Maintenance System

26

Commercial Hardware -1

UNO-2160 Watchdog Agent® System

High Performance and Rugged

– Fanless Celeron 400MHz CPU– 256/512MB SDRAM

9 30VDC i t

Non-volatile Storage Media– CompactFlash– 2.5’’ HDD– Battery Backed RAM

Built-in Comm. Interfaces– 2x Ethernet ports

– 9-30VDC power input– -10 to 50C operating temp– Hardware Watchdog Timer

– 4x Serial ports– 2x USB ports– 1x Printer port

Expansion Capability– 1x PC Card slot (PCMCIA)– 2x PC/104 slots

Robust Embedded OS– Windows CE.NET– Windows XP embedded

Commercial Hardware -2

I/O– Analog Input/Output,

Digital Input/Output– (Low / high speed,

simultaneous /

CompactRIO reconfigurable embedded system

simultaneous / multiplexed)

Real-time controller– 200 MHz Pentium Class– 64 MB or 512 MB FlashReconfigurable Chassis– 4-slot or 8-slot 1 M Gate or

3M Gate FPGAConnectivity– Standard connector

depends on the module– Options: Strain Relief

Connector Block, Custom Cable, etc.

Page 27: Intelligent Maintenance System

27

Watchdog Agent Tools on LabVIEW

IMS Watchdog Agent® in LabVIEW

LogisticRegression

PerformanceAssessment

FeatureMap

CMACPattern Match

PerformancePrediction

Match Matrix

NeuralNetwork

HealthDiagnosis

HiddenMarkov Model

BayesianNetworks

Page 28: Intelligent Maintenance System

28

IMS Plug-n-Prognose System

Watchdog Agent®

Toolbox Hardware Platform

+Robust Health

Information

Radar Chart for Components Health Monitoring

Confidence Value for Performance Degradation MonitoringData

55

Health Map for Pattern Classification

Risk Radar Chart to Prioritize Maintenance Decision

Watchdog Agent®

Platform

Embed Prognostics in Controller

Page 29: Intelligent Maintenance System

29

Validated Hardware

• Allen-Bradley ControlLogix/1756 Rack withLogix 5550

• Special Application Module for ControlLogix – Intel Pentium 266MHz CPU, 128MB of RAM and

Windows XP Professional as operating system– Communication between 56SAM module and

processor is realized over the backplane of the rack

57

Smart Machine Testbed(IMS/TechSolve in Cincinnati, OH)

Page 30: Intelligent Maintenance System

30

Smart Self-Maintenance Machines

MTHI0. 690

MTHI0. 925

MTHI0. 560

MTHI MFPIConstraints MFHI

ProductM1

MTHI0. 751

1. Schedule2. Budget3. Quality

Tolerances4. Etc…….

0.745

Product Quality

Downtime

Maintenance

Flexibility

Readiness

EnergyM2

M3

M4

M5

M6

Prognostics Tools for Vehicle Telematics• Oil pressure and Oil levels

• Fuel level

Od t di• Odometer reading

• Vibration and Acoustic data

• Air Pressure in tire and Flat tire

• Condition of air bags

• Box itself

• Different points on engine and transmission

• Anti brake system

Page 31: Intelligent Maintenance System

31

Historical Working

FeaturesObtain data from the critical locations

Feature extractionSmart Bridge Health Prognostics

Historical Data Embedded

Sensors

Working Condition

In feature spaceNormal Behavior Most

Recent B h i

Health assessment

Behavior

00.20.40.60.8

1

Visualization by performance chart

Beam 1

Truss 1

Beam 2

Truss 2

Support 2

Support 1

Future Combat Systems (FCS) and Needs

Condition-based NeedsIMS Smart Watchdog Agent® Maintenance Platform

PrognosticsRemote diagnostics

L i ti S h i tiLogistics Synchronization

Page 32: Intelligent Maintenance System

32

Self-Maintenance Systems

ModelExpertKnowledge

Performance

Degradation

Model Knowledge

Prognostics

Assessment

Diagnostics

IMSWatchdogAgent™

Sensors

Degradation Information

• Near-Zero-Downtime Service• Autonomous Service Request for Spare Parts

• Asset Optimization

DecisionSupport Tool

Monitoring of Standard Components, e.g.,

Components, e.g.,

Components, e.g.,

Page 33: Intelligent Maintenance System

33

Outline

• Business and Productivity Models• Prognostics Design: Methodology and Tools• Design of Prognostics Platform for Various

Applications• Design of Smart Service Systems

Service Needs for Tomorrow:

• Service is a Customer-Intensive SystemService using a System Instrumentation (Smart• Service using a System Instrumentation (Smart Agent)

• Service through Smart Operation Analytics• Service for Knowledge Management Business• Service to Avoid Potential Issues for Customers.

J. Lee, NSF Symposium on Technology Management in the Service SectorPortland, OR, August 5, 2007 (Co-Sponsored by IBM Service Science)

Page 34: Intelligent Maintenance System

34

Transformation of Service Intelligence

Infotronics Technologies

Synchronized Value Chain Flow

(Cash Flow)

ConclusionsConclusions

Page 35: Intelligent Maintenance System

35

Our Approach Our Approach Value FlowValue Flow

Risk

Validation Company

IMS Platform

DecisionSupport

Watchdog Agent™

IMS Projects Testbeds

ContractsMembership

Research Tools

Validation(Shared Tools & Knowledge)

p y

Designated Projects

and

Implementations

Support Tools

Value

CompanyMembers

Membership and Partnerships

Membership Value through Leveraged Partnerships

Testbeds andProjects

IMS Methodologies

Testbeds andProjects

25:1

IMS Methodologies and Enabling Technologies

Page 36: Intelligent Maintenance System

36

RISK&COST

Today

IMS Value Proposition –Rapid Prognostics Design and Deployment

TIME

RISK&COST

Saving

Value

Risk and Cost Reduction Strategy TIME

IMS Global Research PartnershipsLulea UnivSweden

EPFLSwitzerland

TU-BerlinGermany

Univ. of Toronto,Canada

Shanghai Jiao Tong Univ., China

Beijing Univ of ChemicalS&T (BUCT)

ITRI & PMC

Univ. of TokyoJapan

CRCIntegrated Asset Eng.QUT, Australia

SIMTechSingapore

SENAI/UFRGSBrazil

UTTFrance

Cranfield UnivUK

H.K.PolyU

Taiwan

Page 37: Intelligent Maintenance System

37

Thank You Thank You !!For More InformationFor More Information

Please visit:Please visit:www.imscenter.net

&www.dominantinnovation.com