Dr. Diego Galar - SUECIA
Dr. Diego Galar - SUECIA
Dr. Diego Galar - SUECIA
Luleå
170 km south of the
Luleå Arctic circle
Sweden Finland
Norway
Denmark
Luleå University of Technology
44 %
Education Facts about
Luleå University of Technology
Founded 1971
Turnover EUR 160 million
90 MEUR 70 MEUR 16 000 students
1,570 employees
180 professors
530 teachers & researchers
56 % 580 PhD students
Research Space
Main Campus 70 research subjects in
Art&Media
9 research areas
Wood
25,000 people have graduated from
LTU, of which 13,000 work in the region
Research Programs
Structure of Operation and Maintenance
(O&M) research:
A multidisciplinary approach
RAMS: Reliability Availability
PHM
Maintainability Safety
CBM ILS: Integrated Logistic Support
MPM
RAMS
SKF-UTC and SMART bearings
SKF, groundbreaking intelligent bearing
technology
Maintenance thoughts
Performance Lack of
contract Budget
Decreasing
knowledge
Output
Maintenance Lack of
Contracts information
Others’ thoughts about
maintenance
What will be the cost How do we ensure the
of a better contracted
performance? performance?
Why is
maintenance so What are the effects
expensive? of cutting budgets?
What do we have
to pay for 1% What are the best
more availability? actions to be taken to
improve performance?
Maintenance manager
Maintenance actions and
consequences
What is the (quantitative) relation between
maintenance(cost) as input and performance
of infrastructure as output ?
10
How to answer that question?
Looking for a systematic approach to link
maintenance activities to performance.
• Based on risks
• Usable for evaluation and planning.
Risk Based Approach
Device Operator
Malfunction Error
Inspection
The world is smaller and flatter
and is getting more complex
ERP
Customer
Logistic
ERP
CI
Warning
CM indicates need for a system
shutdown based on CI but….
SMART SYSTEM
ERP
SCENARIO
RUL
SCENARIO 2
N
RUL
RUL
Four ways to classify RISK
decisions
Optimum LCC
Versus
mission
accomplishment
Smarter Assets
enable
new business models
based on RISK
RISK based maintenance
require SMARTness….
Resources
allocated
Maintenace SCENARIO
policy 1 1
Capacity or
RUL other KPIs
for MP1
Maintenace Resources
allocated SCENARIO
policy 2 2
RUL Capacity or
other KPIs
for MP2
Maintenace Resources
allocated
SCENARIO
policy n N Capacity or
other KPIs
RUL for MPN
SMART bearings, the upgrade
towards SMART assets….
Environment
Enterprise System
Feedback
Inputs Outputs
Maintenance
Labour force
System
Load
Grease
RUL
Adjust Remaining
Calibrate Useful Life
Inspect
Repair
Replace
Asset with SMART bearings will get
SMART
Environment
Enterprise System
Feedback
Inputs Outputs
Maintenance
System
Materials Capacity
Labour Quality
Spares Maintainability
Tools Availability
Money Reliability
Technology Safety
Information Cost
Facilities
SMARTness will provide RISK management
Fleet System
Level Level Subsystem Component
Integrated Systems Health Management Continuous
Process Research
Prognostics Improvement
Project future health of the
system, taking into account
estimates of past and
future operation profiles
Remaining
Useful RUL
Life
Logistics
Diagnostics/
Condition
Health
Monitoring/
Assessment
Determine current State Detection
health of system or Compare features against
subcomponents expected values or
operational limits
Inspection Data
Product support
Data
Manipulation
Pre-Processing
Sensor Feature Extraction
Data Signal
Characterization
Inspections Maintenance
Data Recorders
Let us close the loop and mitigate the
risk
In the beginning…
• Robots were designed to be a
replacement for repetitive,
labor oriented tasks
• Humans are slow, imprecise,
prone to injury and have
physical and emotional
limitations
• Machines are fast, precise,
durable and their limitations
make them ideally suited to
repetitive/dangerous tasks
Industrial Robots
• The first modern industrial robots were
probably the "Unimates", created by George
Devol and Joe Engleberger in the 1950's and
60's. Engleberger started the first robotics
company, called "Unimation",
and has been called the
"father of robotics."
Laws of Robotics
• Asimov proposed three “Laws of Robotics”
Legged Robot
Wheeled Robot
What is a Robot?
Autonomous Underwater Vehicle Unmanned Aerial Vehicle
What Can Robots Do?
Decontaminating Robot
Cleaning the main circulating pump housing in the
nuclear power plant
Smarter assets: Systems more efficient,
productive and responsive.
= SAFER and LESS RISKY
NEW
SUSTAINABLE
INTELLIGENCE
AND BEYOND
SMART
DYNAMIC
WORK
ASSET MANAGEMENT
+ +
INSTRUMENTED INTERCONNECTED INTELLIGENT
How should these SMART
assets be?
Instrument the assets
A billion transistors per human being on the planet
Interconnect them
A trillion devices all giving off data – the ‘internet of things’
34
A SMART bearing proposal…..
SCADA Supervisory Control And Data Acquisition
ERP Enterprise Resource Planning
CMMS Computer Maintenance Management Software
FEATURE FAULT
DATA COLLECTION SIGNAL PROCESING
EXTRACTION CLASSIFICATION
Required functions
O R
R
R S
K
MANAGERIAL DATA
ACTUATOR
CONDITION
EVALUATION
SMART BEARING
What does it mean to become Smarter?
Measuring, Monitoring, Modeling and Managing
Data Modeling
+ Analytics Data modeling and
analytics to create
insights to optimize
Virtually all things are becoming smarter decisions
INTELLIGENT Visualization
+ Decisions
What does it mean to become Smarter?
Real Time
Data Integration
Sensing Metering
Real Time
+ Historical Data
RISK ANALYSIS
FEATURE FAULT
DATA COLLECTION SIGNAL PROCESING
EXTRACTION CLASSIFICATION
Required functions
O R
R
R S
K
MANAGERIAL DATA
ACTUATOR
CONDITION
EVALUATION
RISK REDUCTION RISK ASSESSMENT
SMART MACHINE
First challenge: Instrumented and
connected
Value is created by interconnection
and new services
Traffic Services
Vehicle Services Police/Emergency Personal Services
Vehicle Monitoring Traffic Congestion
News
Roadside Assistance Weather
Entertainment
Advanced Diagnostics Concierge
Banking
Software Fault Analytics Insurance
Government
Service & Warranty Data (“pay as you drive”)
Utilities
Vehicle to
Vehicle Dealer
Vehicle to
Electric Vehicle/ Roadside
Tolls Hybrid Charging GPS
Our planet is a complex, dynamic, highly interconnected
$54 Trillion system-of-systems (OECD-based analysis)
This chart shows ‘systems‘ (not ‘industries‘) Communication Transportation
$ 3.96 Tn $ 6.95 Tn
Education
$ 1.36 Tn
Water
$ 0.13 Tn
Leisure / Recreation /
Electricity Clothing
$ 2.94 Tn $ 7.80 Tn
Global system-of-systems
$54 Trillion
(100% of WW 2008 GDP)
Healthcare
$ 4.27 Tn
Architecture
FEATURE FAULT
DATA COLLECTION SIGNAL PROCESING
EXTRACTION CLASSIFICATION
Required functions
O R
R
R S
K
MANAGERIAL DATA
ACTUATOR
CONDITION
EVALUATION
SMART BEARING
46
SCADA Supervisory Control And Data Acquisition
ERP Enterprise Resource Planning
CMMS Computer Maintenance Management Software
Islands of Information
FEATURE FAULT
DATA COLLECTION SIGNAL PROCESING
EXTRACTION CLASSIFICATION
Historical data
CMMS
DATA
SENSORS ERP
FUSION
SCADA
W DE
O
O R
R
R S
Required functions
K
Control Info.
ACTUATOR
CONDITION
EVALUATION
eMaintenance
system
SMART BEARING
Growing
Dumber
Sensemaking
Algorithms
Time
50
The Way Forward
All Digital
Data
Computing Power Growth
Context
Engines
Sensemaking
Algorithms
Time
Introducing … Persistent Context
Notice Respond
Evolving Maintenance Strategies
Context Driven Prognosis
Prognostics & Health Management (PHM)
Preventive Maintenance
Inspection
Run to failure
RBM
Context
Driven
Prognosis
PHM CBM
Driving Smarter Outcomes…
What is the innovation in SMART
bearings?
Capture Predict
(Risk analysis) (Risk Assessment) Act
(Risk mitigation)
…
Text Risk
Mining Assesment
Data
Platform
Mining
From RISKs event point of view…
Event processing in a nut shell
Event Consumer
Event Producer Event Processor
57
A smarter bearing:
addressing the needs of the industry
A dynamic infrastructure…
REDUCE COST
Containing operational
cost and complexity today…
… Achieving breakthrough
productivity gains.
MANAGE RISK
Addressing today’s
security, resiliency,
and compliance challenges… … Preparing for the new risks
of a more connected and
collaborative world.
No more crystal balls
Sharpening
the crystal ball
I’m running at
I need more 1434 rpm
grease!
I’m 75°C
The load is
455 N
I’m unaligned
My loss rate of
Remaining Useful Life
is 3172 rev/min
I feel fine but you’d
better check the rest of
the machine
Prognostics: Let us be aware of its
limitations