The document discusses the use of the industrial internet to transition from planned maintenance to predictive health maintenance, highlighting the economic potential and benefits of prognostics for industrial equipment. It outlines the role of the Industrial Internet Consortium in fostering collaboration among industries, academia, and government to develop standardized protocols and innovative solutions. Key points include the economic impact estimates and the advantages of using advanced data analytics to enhance the longevity and performance of assets.

1. Using the Industrial Internet to Move from Planned
Maintenance to Predictive Health Maintenance

2. Todays Presenters
Ed Wagner
Vice President,
Customers and Solutions
Sentient Science
ed@sentientscience.com
Lynne Canavan
Program Director,
Industrial Internet Consortium
canavan@iiconsortium.org
Stephen Steen
Head of Industrial Internet
Services,
Sentient Science
ssteen@sentientscience.com
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

3. Overview
1. Overview of the IIC and the business problem IIC solves
2. Sentient on the Industrial Internet – Prognostics and Making
Lifing Assessment for Industrial Equipment
3. Renewables Business Case and Outcomes – Moving from
PPM to PHM
4. Questions and Summary
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

4. The Industrial Internet
Industrial Internet:
An internet of things, machines,
computers and people…
enabling intelligent industrial
operations…
using advanced data analytics
for transformational business
outcomes.
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

5. The Industrial Internet is Leading the Next
Economic Revolution
GDP data extracted from the Futurist 2007
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

6. The Impact of the Industrial Internet
General Electric estimates that the Industrial Internet could add $10 - $15 trillion to the global GDP over the next 20 years.
McKinsey Global Institute estimates an annual economic impact of $2.7 trillion to $6.2 trillion by 2025.
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

7. Yet There are Current Roadblocks to
Widespread Adoption
The Industrial Internet:
A $32 trillion opportunity
Connectivity
Technology
Standards
Research Academia
Systems
Integration
Security
Government
Big Data Industries
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

8. The IIC: Things are coming together
Things are coming
together.
Technology
Standards
Research Academia Systems Integration
Security
Government
Big DataConnectivity Industries
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

9. Industrial Internet Consortium
Mission
To accelerate growth of the Industrial Internet by coordinating
ecosystem initiatives to connect and integrate objects with people,
processes and data using common architectures, interoperability
and open standards that lead to transformational business
outcomes.
• Launched in March 2014 by five founding members:
Community.
Collaboration.
Convergence.
The IIC is an open, neutral “sandbox” where industry,
academia and government meet to collaborate, innovate
and enable.
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

10. Our Objectives
• Drive innovation through the creation of new industry use cases and testbeds for
real-world applications;
• Enable interoperability by defining and developing the necessary reference
architecture and frameworks
• Influence the global development standards process for internet and industrial
systems;
• Build confidence around new and innovative approaches to security.
• Facilitate open forums to share and exchange real-world ideas, practices, lessons,
and insights;
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

11. IIC Members
11
IIC Founder Companies
www.iiconsortium.org
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

12. Model-based, multi-physics based
prognostics computational technologies
and services
Our applications help extend the remaining
useful life (RUL) of new and existing
mechanical systems
The newest prognostics health
management (PHM) application for
condition-based maintenance (CBM)
Sentient Science is Based on Three
Fundamental Capabilities
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

13. DigitalClone Prognostic Validations
Rotorcraft
• NASA Spur Gear Validation
• “Large Bearing Company” Bearing Validation
• Apache Spiral Bevel Gear Validation
• Actuator Pump Cam Bearing Validation
• Rotorcraft Super Gear Validation
• Blackhawk Spline Fretting Fatigue Validation
• Navy Elastomeric Bearing Validation
Wind Turbine
• Clipper Wind Turbine GBX Validation
• GE 1.5 Wind Turbine GBX Validation
Automotive, Mining & Others
• Planetary Gear System Validation
• Differential Assembly – Spiral Bevel Validation
• Axle Shaft Bearings Validation
• Automotive Structural Materials Validation
• Reduction/Axle Hypoid Gearing Validation
• Mining Sag Mill Gear Validation
• Hip Implant Wear Validation
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

14. DigtialClone®
Nucleation & Propagation
<12 months1-3 years3-5 years+5 years
Using the IndustrialInternet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

15. Prognostics on the Industrial Internet
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015
Step 1: Predict the performance
and health of your assets,
components and systems from
0-6 years out.
Step 2: Acquire
operational data, per
fielded asset using your
data system and
networks, to tailor service
for each individual asset.
Step 3: Confirm
prognostic accuracy
through advanced
processing, alerts, and
reporting tools.
Step 4: Control or optimize
assets life through, trade-off
studies, automation,
optimization.

16. Serial
Number
Model
Part
Number
Fleet
Model
Sentient - 3rd Generation
Life Model
2nd Generation
Life Model
Process
NO CAPEX
$31,000
8 weeks
$3.5M CAPEX
$323,000
1 year
System Model with 3-5 Top Failure Component Models
Zero
$100 in compute time
How to build Life model Cost & Time Value Prop comparison
$20,500
Interpreting results difficult

17. What’s the Business Problem to Solve
• Owner/Operators of industrial Equipment trying to assess financial
risk and liabilities associated with off warrantee assets
– “How long will it really take me to get usable information regarding my
fleet and individual assets?”
– How long will these assets produce value?
– Will my O&M costs outweigh my financial pro forma?
– How can I improve budgeting over 2-5 years?
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

18. What’s the Business Problem to Solve?
• OEMs of industrial Equipment trying to assess financial risk and
liabilities associated with their warrantees
– How soon can I find out what my real risk is for assets under warrantee?
– What’s the right balance between reconfigure vs. replacement?
– What’s my downside of not providing competitive warrantee services?
– How do I budget to service customers today and into the future?
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

19. Tim
e
Conditio
n
Extend the Remaining Useful Life (RUL) through Prognostics
®
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

20. Manage the Fleet Through Financial
Optimization
Year: 2014
Today
Year: 2011
3.7%
As-Is Failure Profile
Optimized Condition
Reduction in % of
Failures
Sentient’s services predict the
true baseline and impact of
maintenance and operational
changes.
Life extension or increased AEP
by quantifying the impact of:
- Wind conditions
- Lubrication change
- Turbine re-rating
- Uptower replacements
- Remanufacturing
- Automated spare parts
management
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

21. “As Is”
Quantify Impact of: Lubrication change, up-tower component replacement, partial de-rating, etc.
“To Be”
Years Until Gearbox FailureYears Until Gearbox Failure
PercentageofFleetFailure
PercentageofFleetFailure
Linking Fleet Level Performance to
Component Level Performance
Prognostics allows you to quantify “what-if” scenarios to extend RUL
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

22. Business Impact of Prognostics
• Prognostics cut cost of the Big Data need to maintain assets
– Prognostics vs. Diagnostics
• Dramatically Reduces Time to Value for Asset Health Information
– Lifing predictions within 3 months
– Impact on new customer acquisition for OEM
• Battle for the Aftermarket – will the Operator or the OEM dominate
the aftermarket
– Reconfiguration (Owners) vs. Redesign (OEMs)
– “What-if” simulations creating significantly more competition
March 16, 2015
The Financial Impact of Extending Asset Life

23. Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015
Sentient Science and the Industrial Internet Wave
General Electric estimates that the Industrial Internet could add $10 - $15 trillion to the global GDP over the next 20 years.
McKinsey Global Institute estimates an annual economic impact of $2.7 trillion to $6.2 trillion by 2025.

24. Industrial Internet
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

25. Challenge for Wind Owner/Operators
Failures higher than originally planned during site development.
Assumptions used:
– 100 WTG’s
– Loan – 18 yr tenor
 Approximately 50% of the total cash
to equity comes in the last 5 years
10.00%
10.50%
11.00%
11.50%
12.00%
12.50%
13.00%
13.50%
14.00%
14.50%
15.00%
0.5 1 1.5 2
$40,000
$50,000
$60,000
$70,000
$80,000
$90,000
$100,000
20YearIRR%
CashFlowtoEquity,Yrs15-20($,000)
GBX Failure / WTG
Cash Flow to Equity Years 15-20 20 year IRR
Budgeted Actual
*Owners are looking to gain
control of cost
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

26. Original Pro Forma O&M Plan
Planned Preventative Maintenance (PPM)
Pre-planned based on usage or time.
Pro: Almost always
ensures proper operation
Con: Expensive and
doesn’t address outlier
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015
Cannot predict future risk or solutions to reduce it

27. Improved O&M - First Attempt
Condition Based Maintenance (CBM)
Based on known need for maintenance.
Pro: Reduced cost &
improved reliability
Con: High upfront cost,
monitoring cost &
produces false positives
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015
Can only predict short term failure, cannot provide
solutions to reduce risk

28. O&M Solution – Path Back to Pro Forma
Prognostics and Health Management (PHM)
Predicts future failure to plan and mitigate risk.
Benefits:
1. Little to No Monitoring
2. Less Sensors
3. Remaining Useful Life (RUL)
Extension
4. Risk Mitigation
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015
Can predict risk years into the future and provide ways
to reduce risk through life extension

29. DigitalClone Live
Asset Ranking Default Most-at-Risk, Sortable, Filterable …
Asset Map View Risk Level,
Interactive, Navigable
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

30. Component Risk Failure risk ranking down to the component
level for O&M planning (e.g. uptower vs crane)
DigitalClone Live – Component Level
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

31. Todays Prognostic Reports
1. “As Is” failure rates at the fleet, site, critical component levels
– Watch lists, borescopeing, inventory and supply chain management
– Budgeting
2. Gearbox replacement schedules for forecasting
3. “To-Be options to extend life:
– Component replacement
– Lubrication and additive impact on life
– Duty cycle changes to improve life
– “What-if” simulations before taking action
You Can “See” What Diagnostics Can Not Yet See
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

32. Asset Prognostics over the Industrial Internet
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015
Predict initial WTG
risk for each
individual asset
Acquire operational
data (SCADA,
vibration, etc.) from
individual WTG
Confirm the WTG
prediction to the data to
ensure it is operating as
predicted
Control and optimize the
WTG through re-rating,
up-tower component
replacement

33. Example Prognostics: ROI Optimization
ROI Optimization All trade-offs based on ROI at the
individual asset level/group, maximizing ROI for the fleet.
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

34. 1.5mw WTG – 500 Units
$2,000,000.00
$7,000,000.00
$12,000,000.00
$17,000,000.00
$22,000,000.00
$27,000,000.00
$32,000,000.00
$37,000,000.00
$42,000,000.00
$47,000,000.00
$52,000,000.00
$-
$5.00
$10.00
$15.00
$20.00
$25.00
$30.00
6 7 8 9 10 11 12 13 14 15 16
AS-IS MW/Hr Cost TO-BE MW/Hr Cost
AS-IS Yearly Cost TO-BE Yearly Cost
MW/Hr
Cost
• Through life extension,
failures rates can be
managed and reduced
• Sentient can provide a
significant improvement
each year to the failure
rates
5 years, $19M savings, 500% ROI
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

35. Business Impact
1. Customer 1: Failure rate reduced by +50%
– Customer budgeted and planned or 20+ gearboxes for 2014
– Through up-tower replacements, derating, etc. actual failures rates reduced to <10
– $5,000,000 cost reduction in budgeting
2. Customer 2: No failures, correctly predicted next gearbox failure
– Was able to plan and prepare for gearbox failure in 2014
– 2-5 year PHM budgetary plan
3. Customer 3: Identified and confirmed 8 gearboxes on top 20 list not
previously known or expected
– 90% correlation with borescope conducted on top 20 gearboxes for exact component
of failure
– Crane and service optimization at $70K per crane service, reduced unplanned
maintenance
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

36. Applying Prognostics on the Industrial
Internet
Supply Side Optimization
Using Prognostics
Inventory and Supply
Chain Optimization
Fleet Process
Optimization
O&M Improvements
SMART
GRID
1
2
3
4
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015

37. Summary
• Sentient’s DigitalClone Live powers the move to Predictive Health
Management (PHM) using the Industrial Internet
– DoD, Industrial, Energy Use Cases
– Demonstrations on request
• IIC is the central organization for pooling technologies, security,
architectures, test beds for learning and leveraging best practices
– www.industrialinternetconsortium.org
– March 24-26, 2015 - Reston, Virginia USA
– July 15-17, 2015 – Niskayuna, New York USA
– September 14-17, 2015 – Barcelona, Spain
– December 7-14, 2015 - La Jolla, California, USA
Using the Industrial Internet to Move From Planned Maintenance to Predictive Health Maintenance
March 11, 2015