IoT Analytics Martin Keseg Enterprise Account Manager 26/10/2017
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
IoT Analytics
Martin KesegEnterprise Account Manager
26/10/2017
What is IoT Analytics?
Raw Goods Equipment Temperature/Pressure
Equipment Vibration
Equipment Utilization Trip Distance Customer Behavior
SpeedIdle Time Inclement Weather
Equipment Vibration
IoT is a Big Data Problem
Equipment Temperature
Customer Behavior
Equipment Pressure
SpeedEquipment Utilization
Trip Distance
LARGE VARIETY OF SENSOR DATA
LA
RG
E V
OLU
ME
OF
DA
TA P
ER
DA
Y (
> M
ILLI
ON
)
Contextualize the Data
ERP
CRMAsset
Management
STRUCTURED DATA
Sensor Data
Geo-Location DataMachine Data
Image Data
Video or Voice Data
SEMI-STRUCTURED DATAUNSTRUCTURED DATA
What is the solution?How to contextualize all these data?
Contextualize the DataSTRUCTURED DATASEMI-STRUCTURED DATAUNSTRUCTURED DATA
SIMPLIFIED BY PENTAHO IOT ANALYTICS
Ingest Process BlendData Prep
Action
Machine Learning
ERP
CRM
Asset Management
Sensor Data
Geo-Location Data
Machine DataImage Data
Video or Voice Data
6
IoT Analytics Value Tenants
321
Sense & Capture Integrate & Blend Infer & Act
7
IoT Analytics Value Tenants
1
Sense & Capture
1
Asset Mgmt
Register Sensors
Model Assets
Store Sensor Data
Stream Sensor Data
8
IoT Analytics Value Tenants
2
Integrate & Blend
2
Data Refinery
Validate and Cleanse
Process & Blend
Stream Processing
Rules Engine
9
IoT Analytics Value Tenants
3
Infer & Act
3
Analytics
Machine Learning
Report & Alert
Workflow Integration
Business Outcomes
10
3 IoT Platform Modules
1
Asset Mgmt
Register Sensors
Model Assets
Store Sensor Data
Stream Sensor Data
2
Data Refinery
Validate and Cleanse
Process & Blend
Stream Processing
Rules Engine
3
Analytics
Machine Learning
Report & Alert
Workflow Integration
Business Outcomes
IoT Platform Demo Scenario
12
Big Fleet (222 Vehicles)
24
46
152
Web-Based Fleet Management Platform
13
Asset Model Types
14
Asset Models
15
Hierarchical Vehicle Modeling
Store models and sensor data
Asset ModelSensor DataUtility Vehicle Asset
Air Pressure
Axle Vibration
Lights
Load Weight
Movement
Temperature
Sensor Data
Journey
StreamBlend
Infer
Sense
Inspect
Embed & Integrate
Store
16
Adding Context to Sensor Data
Vehicle Location
•GPS
•Lat / Long
•Mapping
•Movement
Vehicle Profile
•Make
•Model
•Mileage
Operational Systems
•Maintenance History
•Maintenance Schedule
•Service Centers
•Parts Ordering
•Parts Inventory
Sensor Data Contextual Data
Business Outcomes
• Real-Time Fleet Status and Health
• Repair Recommendations
• Optimized Maintenance Scheduling
• Automated Parts Ordering
IoT Data Refinery
Sensor Data
Journey
StreamBlend
Infer
Sense
Inspect
Embed & Integrate
Store
Pentaho’s IoT Analytics Workflow Orchestration
IOT Data Refinery
Analytic Database
Pentaho Analyzer
Sensor
Traditional Data
Pentaho Data
Integration
Pentaho Reporting
MSG QueueKafka, JMS, MQTT
Machine Learning
R, Python, Weka
Stream Feedback Loop
LOB Applications
Embedded
Pentaho Data
Integration
Why Pentaho?
Industrial IoTwith Hitachi
Expertise in Machine Learning
Solve Big Data Problems
for IoT
Telematics Analytics
Customer Experience
Trains as a Service
Driver, Equipment and Fuel Analysis
Categorize Customer
Preference
Predictive Maintenance
Reduced Down Time and Cost
Savings
Customer Retention & Upsell
Multi-million Dollars Maintenance
Savings
IOT Transforms Business Outcomes
Business Requirements IoT Analytics Business Outcomes
Q & A
Customer Use Case in Detail
Customer Experience – IMS
Challenges
• Create custom dashboards
• 1.6 Billion data points/day
• Predictive behavior patterns
Solutions
• Seamless integration to IMS data sources
• Embedded IoTAnalytics
• Data-driven insight and predictive analytics
Benefits
• Retain customers
• Reduce driver claims
Predictive Maintenance – Hitachi Rail Europe
Challenges
• 3.6 million data points per second
• Correlating multiple data points
• Visualization in multi-tenant
Solutions
• End-to-end Big Data platform
• Scale to data growth
• Predict operational events
Benefits
• Millions maintenance savings
• More reliable andcost effective
• Better service delivery
Customer Experience – Veikkaus
Challenges
• Stream customer behavior –20K terminals
• Blend structured and unstructured data
• Data from other gaming platform
• Slow reporting
• Casino competition
Solutions
• Open architecture
• Tight Integration to Cloudera HDFS
• Professional services
• 24/7 support
Benefits
• Marketing
Timely and accurate data
Customized offers
• Boost sales
• + Customer experience
• Identify gambling addiction
Telematics Analytics – Caterpillar
Challenges
• Integrate sensor data and other data
• Data integrity was in question
• Dashboard development
Solutions
• Connected all data sources
• Ingest, process and blend all sensor data
• Process high volume of data
• Faster delivery to market
Benefits
• Adjust power to operate the generators – $650K savings
• Reduced ship’s hull build-up – $800K fuel reduction cost per ship
Thank You
Related Documents
