How do we do we make decisions at the speed of business? Traditional supply chain processes are batch, and out of cadence with business. How do we rethink these processes to have the right data available when we need it. In this presentation, we discuss the inclusion of streaming data in supply chain visibility. It is not sufficient to ask the question of "Where is my stuff?" without the opportunity to use the data in better decision making.
Presentation at the October Scope Event on Internet of Things
1. Global Supply Chain Visibility
from IoT and Machine Learning
Solutions
SCOPE SUPPLY CHAIN CONFERENCE
OCTOBER 9-11, 2016
MARRIOTT MARQUIS SAN DIEGO MARINA
2. Jim Hayden
Senior Vice President
Savi Technology
Lora Cecere
Founder and CEO
Supply Chain Insights
3. Can Your Supply Chain Operate at
the Speed of Business?
LORA CECERE
FOUNDER AND CEO
SUPPLY CHAIN INSIGHTS
15. Global Supply Chain Visibility from
IoT and Machine Learning Solutions
JIM HAYDEN
SENIOR VICE PRESIDENT
SAVI TECHNOLOGY
16. Internet of Things “IoT”
Sources: Cisco, IBM, IDC
Unstructured data from sensors is a much more salient
feature of what is being called big data
The “sensorization” of society is unlocking for the
first time in human history, the potential to gather
in real time enormous amounts of data and details
about almost everything.
With the added complexities introduced by new data
sources (such as real-time events and sensors) and
new types of analysis new opportunities will emerge
to build business value.
SENSORS BANDWIDTH PROCESSING
IoT Enabler:
Average Costs Over 10 Years
IBM to Invest $3 Billion in
Sensor-Data Unit New business to help
customers gather and analyze the flood of data from
sensor-equipped devices
90%
of data generated by connected
devices is never analyzed
18. Which vessel is my container on?
How long has my shipment been
sitting at the port?
Are my raw materials going to get
here in time?
What temperature and humidity
are my products at right now?
19. This is the real value of IoT
for Supply Chain.
20. Leveraging All Data Sources …
IoT Unstructured Data
Sensors
Mobile Device Apps
Fleet GPS Data, Telematics
21. Ocean Visibility – A Particular Challenge
Sensors using terrestrial mobile network
◦ Typically transmitting infrequently
◦ Time, location, environmental measures
◦ Only reliable close to land
Sensors using satellite network
◦ Typically transmitting frequently
◦ Time, location, environmental measures
◦ More expense, only reliable if container is in line of sight
AIS data for tracking
vessel MMSI="413811000" TIME="2011-04-12 10:40:27
GMT" LONGITUDE="118.44586666667" LATITUDE="38.87483333333
3" COG="356" SOG="0.1" HEADING="116" NAVSTAT="1" IMO="
9118824" NAME="JIN HAI
XIANG" CALLSIGN="BVKU" TYPE="70" A="197" B="27" C="20"
D="12" DRAUGHT="7.5" DEST="CAOFEIDIAN" ETA="04-10
07:00"
AIS data feeds using satellite network
◦ Typically transmitting every 6 minutes
◦ Time, location, destination
◦ Most reliable, sometimes turned off (avoiding piracy)
23. IoT Platform – Core Capabilities
Sensor
Communication
and Management
Data Storage/
Cleansing and
Mapping
Presentation/
Business
Connectivity
Open Source Data
Enterprise Data
Data Analytics/
Deep Learning
structured
data
semi-structured/
unstructured
unstructured
data
“things” thing connectivity data management machine learning presentation
25. Machine Learning on IoT Data
In 1959 Arthur Samuel defined machine learning as
a “Field of study that gives computers the ability to
learn without explicitly being programmed.”
Requires Data Scientists
26. Models for Predicting Arrival Time
ETA Late confirmed
Shipment on time
ETA Late warning
Machine Learning generates most accurate ETA forecasts
29. Why Should You Care?
Category Benefits
Detention and Demurrage Costs Cost reduction due to ability to capture actual time at a
location
Cross-Docking Costs Improvement due to more accurate ETA of inbound
shipments
Late Fees and Expediting Fees Decrease due to improvements in estimating shipment
times
Carrier Performance Decrease in lane holding time and improve carrier
compliance through accurate data and analytics
Inventory Value of inventory on hand reduction due to decreased
lead time and variability
Lost Sales Percent reduction of lost sales due to stock outs because
of fewer missed and late deliveries
Theft and Damage Percent reduction in theft, loss and damaged goods due
to real-time visibility
Logistic Planning Percent reduction of logistics planning and forecasting
personnel due to automated modeling
30. Savi Supply Chain Innovator
revolutionary
technology
delivered as
purpose-built applications
that solve strategic supply
chain challenges
Industry awards
and recognition
Supply Chain
Analytics Pioneer
leveraging M2M data &
the Internet of Things
to provide provide
logistics insight
World's Largest
Asset Tracking
System powered by Savi
RF-ITV
award-winning
solutions track
more than
in goods annually
Driving supply
chain success
with leading Global 1000
organizations
IP portfolio
currently hold more than
60 patents
related to logistics,
IoT, and in-transit
visibility