In this webinar, experts from VoltDB and Teradata discussed top-of-mind topics to help you gain immediate traction in your IoT architectures. Cheryl Wiebe, Partner, Analytics of Things, Think Big, A Teradata Company and Dennis Duckworth, Dir. Product Marketing, VoltDB will discuss the most impactful IoT use cases and business models, and key technology considerations to ensure effective analytics against IoT data. Finally they will outline real world IoT customer case studies showcasing break-through value. To view the entire webinar, click here: http://learn.voltdb.com/WRTeradataIoT.html
3. 3
Internet of Things now
worth 14.4 trillion dollar up
from 35 billion dollar.
Is amazing Compound
Hype Cycle Growth Rate
(CHCGR) of 41143%
Big Data Borat @BigDataBorat 18 Mar
4. 4
• Smart Meters
• Digital Oil
Field
• Delivery
• Refinement
• Wind/Solar
Management
• Lights Out
Production
• Smart
Warehouses
• Smart Cities
• Waste
Management
• City Works
• Sowing and
Harvesting
• Yield
Prediction
• Plant Disease
Diagnostics
• Delivery
Efficiency
• Driver Safety
• Truck
Maintenance
• Insurance
Based on
Driving
• Market
Predictions
Based on
These Other
Markets
• Smart
Hospitals
• Smart Home +
Healthcare
Possibilities for IoT in Your Industry
Gartner: Internet of Things: The Foundation of the Digital Business, Jan 6, 2016“Analytics are essential to the success of IoT systems. They are arguably the
main point of the IoT as they support the decision-making process in operations
that are created in business transformation and digital business programs.”
Roy Schulte and Rita Sallam,
Three Best Practices for Internet of Things Analytics, October 23, 2015
5. 5
Two Major Subsystems
Operations of Things Analytics of Things
Things
Gateway
The Edge
Networks
Analytics
Data Center
Local | Cloud | Hybrid
6. 6
IoT data introduces new complexities
Data management and Analytics must adapt
Key Issues:
• The velocity and volume of the data may be huge
• In some cases, most of the data is unimportant, or redundant
• In some cases there will be a need to increase/decrease collection
• Some data is clearly erroneous
• May need intelligent processing at the edge
7. 7
Two Types of IoT users
Asset Maker / Seller Owner/ operators
Focus
External…the customers
using their products
Internal…their own
operation
Examples
Large Equipment, Auto,
Aerospace
Utilities, Hospitals, Plants, Oil
Exploration / Refining, Cities
Goal
R&D, warranty,
Product as Service
Improve own operations,
product
Analytics at Edge
Data Compression;
Anomaly detection
Fleet coordination; policy
enforcement
8. 8
Use cases for the Asset Maker / Seller of Heavy Equipment:
Condition Based Monitoring & Maintenance
Improving field services for some of the
most complex, transportation and
heavy asset companies out there
9. 9
• Data Collaboration
– Engines
– Aircraft
– Operator
• Improved Maintenance
– Condition-Based
– Predictive
Product as a Service
• Power by the Hour
– Pay for usage
– Known cost projection
• Pay when it works, not breaks
– Supplier/Operator incentives
aligned
• IoT enables model in lower cost
businesses
Commercial Airline
Services
11. 11
Semi Conductor
Owner/Operator of the Manufacturing Site or Campus
Operations Performance Optimization
Helping high tech, discrete and process manufacturers reduce
equipment downtime, plant throughput, and available-to-promise
Paper / Cellulose Processed Food
Consumer Elec.
Contract Mfg
12. 12
Need to analyze a
constantly changing
set up >20,000
attributes to identify
yield loss factors
required flexible
schema AND large
matrix math
Data prep
• Ingest raw data
using JSON
• UnPivot
Sensor Data
• From Semi-
conductor Probes
(>20,000
variables)
• Principal components
analysis
• Linear Regression
• Embedded Visualization
13. 13
Parse machine logs, to
identify significant
machine events.
Pattern matching helps
predicts machine failure
Affinity analysis predicts
likely replacement parts
• Npath Pattern
detection
• Affinity analysis
• Visualize
Process Automation Equipment
• Line equipment; Sensor logs