Bryan Worth – SAP Solutions Engineer, Aerospace and Defense
April 8, 2014
SAP Federal Forum Unlocking Dark Data from today’s modern Weapons Systems
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
Value of Dark Data in Maintenance
Current Data trend across many industries
Proliferation of OEM Health Management Data
Unscheduled Failures
Preventative On-Condition Maintenance
Cross platform coverage
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
Predictive Maintenance is key to improve readiness within
today’s budget
Early Warnings & Recalls
Warranty Cost Management
Spare Parts Optimization
Monitor
Predict
Act
Onboard Health & Prognostics
Warranty and Claims
Field Service Data
Call Center Data
Engineering BOM
R&D Knowledge Base
Outcome Based Contracts
Other 3rd Party Data
Structured
Unstructured
Design Improvement
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
Doing Predictions basically is a Four Step Approach
Historic data is used to learn. These leanings are used to create prediction models.
These models can than be applied to current sensor data. They need to be
systematically controlled & maintained to ensure best possible results.
Historic Data Prediction
Models
Current
Sensor Data
Control &
Maintenance
Technology Enablers
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
Why now is the time to gain value from “Dark Data”
Database
Predictive
Visualization
Mobility
Cloud
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
.
INNOVATIONS PREVIOUSLY UNFEASIBLE
• Real-time genome analysis
• Instantaneous fraud detection
• Predictive maintenance
• Optimize procurement, manufacturing, transportation
• Real-time MRP with instant re-planning
SIMPLICITY PREVIOUSLY UNACHIEVABLE
• Transactions and analysis in one system
• Efficiently analyze structured and unstructured data
• Hardware cost savings
• Your choice of cloud deployments
SAP HANA In-Memory
Transaction & Analysis
directly In-Memory
VALUES PREVIOUSLY UNATTAINABLE
• Iterative period end closing
• Cash forecasts/management
• Real-time offer calculation
• In-moment sales forecast
• Self-service apps with instantaneous response
• Interactive POS data analysis
SAP HANA
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
SAP Lumira Visualization
Iteratively discover data by building
visualizations
Synthesize and transform data the way you want it
Acquire data from corporate
& personal sources
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
SAP Lumira Example
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
SAP InfiniteInsight: Do More with More Data Quickly
Simplicity: make predictive analytics accessible to non data scientists
Flexibility: automate data preparation, modeling, and production deployment
Speed: build sophisticated models in hours/days instead of weeks/months
Lower TCO: reduce cost by shortening modeling cycles and eliminating manual errors
Scalability: easily scale to thousands of variables and big data volume
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
InfiniteInsight model to predict engine over-temp
Running Model
Reports
Defined model continuously
searching for commonalities to
for failure prediction
Modeling tool
identifies data
correlations
for condition
outliers
Engine failure
data from
fault mode
“High TGT”
Fleet data from
onboard
systems sent to
Hana Platform
SAP HANA In-Memory
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
User Interface – Mobile Device
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
User Interface – Dashboard
How to get started
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
Customer gets
license on
prototype code
for go-live
Idea / Idea Pool
Problem validation
Early Prototypes
Co-Develop Cust. Dev.
Project
Customer Project
Stop
SAP’s Predictive Co-Innovation Process
• Refine
understanding
of customer
problem
• Discuss
possible
solutions
• Technical
feasibility
• Customer
presents a
problem it
has been
facing to
SAP
• SAP
innovations
could
address the
problem
• First
prototypes
(e.g. paper,
then UI
Mockups)
• Decision to
extend project
• Co-developing
prototype for
customer-specific
challenge
• Tests at customer
site with customer
data
• SAP ships pre-
versions for
customer testing
• Each party covers
its own cost
• Projects results
open
P R O T O T Y P E &
V A L I D A T E
C U S T O M E R S P E C I F I C
Legal
coverage: FBA
Collaboration Agreement /
TEA FBA SEAP / EULA / CD-Contract
D E V E L O P
Go / No-Go Decision
N O F O L L O W - U P
I D E A T E &
S E E D
~ 3-5 months
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
US Aeronautic Manufacturer
SAP InfiniteInsight and HANA have been
used to:
• Detect anomalies/outliers against telematics
data
• Identify planned/unplanned maintenance using
HANA Text Analysis
• Detect failures (unplanned maintenance) for the
next coming days
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
Deere & Company – Agriculture Machinery Manufacturer
• JOHN DEERE went live with a custom development solution on SAP HANA for
Emerging Issues Detection.
• The objective was to identify emerging issues before they have a significant impact
in terms of warranty claims. They achieve it by analyzing telematics data, detecting
potential issues, relating them to service and warranty data and shortening the
detection-to-correction cycle.
• In turn, this would improve product quality and reliability, enhance JOHN DEERE’s
brand reputation and build customer loyalty., which will then not only lead to
reduced warranty costs and better profit margins but also repeat business.
© 2014 SAP AG or an SAP affiliate company. All rights reserved.
Questions?
Contact information:
Bryan Worth
Senior Solutions Engineer - Regulated Industries, Aerospace and Defense