2. AGENDA
What is IoT?
What is IIoT
What is AoT
How IoT Analytics is Different?
Case Studies
Data Science (necessity and importance)
Analytics Way Forward
Technology Stack for Analytics
Questions & Answers
4. BASIC DEFINITION
• Forbes : This term refers to devices that collect & transmit data via the
internet.
• SAP : More than 50 billions objects will connect to internet and this
connection is called as IoT. This “things” talk to each other,
collect streaming data and insights.
• Cisco : The IoT links smart objects to the Internet. It can enable an
exchange of data never available before, and bring users
information in a more secure way.
• Wiki : The Internet of Things (IoT) is the network of physical
objects/devices that enables these objects to collect and
exchange data.
5. THINGS IN INTERNET OF THINGS
• Objects
• Machines
• Appliances
• Building
• Vehicle
• People and many more…
6. WHAT IS INTERNET OF THINGS!
6
Application
Server
Things
Web
Service
s
8. M2M
• Machine to machine refers to direct communication between
devices using any communication channel, including wired and
wireless.
• Machine to machine communication can include industrial
instrumentation, enabling a sensor or meter to communicate
the data it records.
12. MARKET DRIVERS & BARRIERS
Four Market Drivers
•Expanded internet connectivity
•High mobile adoption
•Low-cost sensors
•Large IoT investments
Four Barriers
•Security concerns
•Privacy concerns
•Implementation problems
•Technological fragmentation
13. SIZE IN MARKET
$6
TRILLION
INVESTED
$6 trillion will be invested
on IoT solutions over the next five years 24 BILLION
There will be 24 billion IoT devices
installed by 2020
$13
TRILLION ROI
Total investments over the next five years
will generate $13 trillion by 2025
14. ENTITIES USING IOT ECOSYSTEM
Consumers
5B Devices Installed By
2020
Governments
7.7B Devices Installed By
2020
Businesses
11.2B Devices Installed
By 2020
$3B Spent (2015-2020)
$2.1B Spent (2015-2020)
$900M Spent (2015-2020)
$7.6B ROI (2015-2025)
$1.4B ROI (2015-2025)
$4.7B ROI (2015-2025)
15. VERTICALS UTILIZE IOT ECOSYSTEM
Transportation
Manufacturing
Connected Homes
Agriculture
Oil & Gas, Mining
Utilities
Infrastructure
Health Care
Many more….
16. INDUSTRIAL INTERNET OF THINGS (IIOT)
• A universe of intelligent industrial products, processes and services that
communicate with each other and with people over a global network -
Accenture
• The Industrial Internet of Things (IIoT) is the next wave of innovation
impacting the way the world connects and optimizes machines. The IIoT,
through the use of sensors, advanced analytics and intelligent decision
making, will profoundly transform the way field assets connect and
communicate with the enterprise – EDN Networks
• The Industrial Internet of Things is an evolution of existing technologies that
enables end users to improve processes, drive productivity, and maintain an
edge in our increasingly competitive global economy - Kepware
21. INTRODUCTION OF AOT
Connected
Devices
Embedded Analytics Smart Devices
Analytics enables to make predictions and send alerts/notifications from
streaming data(real time sensor data) using automated analytics platform
Analytics decode and transform the continuous flow of M2M data into
value-added information
34. HOW TO DIGEST THIS HUMONGOUS DATA
FOR DATA INSIGHTS
CONFIDENTIAL & PROPRIETARY DOCUMENT
• Performance Analytics
• Real Time Benchmarking
• Trend Analysis
• Adaptive Machine Learning Algorithms
Performance
Analytics
Trend
Analysis
Machine
Learning
Real Time
Benchmarking
40. CHALLENGES WITH ANALYTICS IN IOT
• Data is growing bigger
• Data in motion
• Geographically distribution of data
• New functionality and tools (capability perspective)
44. SOME STATS ON DATA SCIENCE:
• 25% of organizations now have a data scientist on staff.
• By increasing the usability of data by just 10%, the average Fortune 100
company could expect an increase of $2 billion dollars (source: Fathom)
• 1,40,000 to 1,90,000 people with deep analytic skills as well as 1.5 million
managers and analysts will be needed by 2018 to fill jobs in Big Data in US
(source: McKinsey)
• 86% of people are willing to pay more for a great customer experience with
a brand (source: Lunch Pail)
• By 2020 the IoT Will Include 26 Billion Units, Creating New Challenges for All
Aspects of the Data Center (source : Gartner)
46. WHAT IS ANALYTICS?
Maturity of Analytics Capabilities
CompetitiveAdvantage
Raw
Data
Cleaned
Data
Standard
Reports
Ad Hoc Reports
& OLAP
Predictive Modelling
Optimisation
What Happened? – Descriptive Analytics
Why did it happen? – Diagnostic Analytics
What will happen? – Predictive Analytics
What is the best that could happen? Prescriptive Analytics
Sense & Respond Predict & Act
Generic Predictive
Analytics
Source: SAP
47. MACHINE LEARNING
• Machine learning is a subfield of computer science that evolved from
the study of pattern recognition and computational learning theory in
artificial intelligence.
- Wikipedia
• Machine learning is a type of artificial intelligence (AI) that provides
computers with the ability to learn without being explicitly
programmed. Machine learning focuses on the development of
computer programs that can teach themselves to grow and change
when exposed to new data.
- whatis.techtarget.com
48. TYPE OF PROBLEM CATEGORIES
Supervised
Learning
Unsupervised
Learning
49. COMMON PREDICTIVE ANALYTICS
METHOD
• Regression:
Predicting output variable using its cause-effect relationship with input variables. OLS
Regression, GLM, Random forests, ANN etc.
• Classification:
Predicting the item class. Decision Tree, Logistic Regression, ANN, SVM, Naïve Bayes classifier
etc.
• Time Series Forecasting:
Predicting future time events given past history. AR, MA, ARIMA, Triple Exponential
Smoothing, Holt- Winters etc.
50. COMMON PREDICTIVE ANALYTICS
METHOD (CONTD.)
• Association rule mining:
Mining items occurring together Apriori Algorithm.
• Clustering:
Finding natural groups or clusters in the data. K-means, Hierarchical, Spectral, Density
based EM algorithm Clustering etc.
• Text mining:
Model and structure the information content of textual sources. Sentiment Analysis,
NLP
51. BUSINESS APPLICATIONS OF PREDICTIVE
ANALYTICS
Factory Failures
FinanceSmarter Healthcare
Multi-channel
sales
Telecom
Manufacturing
Traffic Control
Trading Analytics Fraud and Risk
Renewable
Energy
Spam Filters
Retail: Churn
53. LEARNING LINKS
• https://www.coursera.org/browse/data-science
• http://www.kdnuggets.com/2015/09/15-math-mooc-data-science.html
• http://www.datasciencecentral.com/profiles/blogs/how-to-become-a-
data-scientist-for-free
• http://www.r-bloggers.com/in-depth-introduction-to-machine-learning-
in-15-hours-of-expert-videos/
• http://online.stanford.edu/course/machine-learning-3
• http://online.stanford.edu/course/mining-massive-datasets
• Kaggle : The leading platform for predictive analytics competitions
• https://www.analyticsvidhya.com/
54. QUESTIONS AND DISCUSSION
THANK YOU SO MUCH FOR YOUR VALUABLE TIME
Mitesh Gupta
Email: mitesh.gupta19@gmail.com
LinkedIn: https://in.linkedin.com/in/mitesh-
gupta-28014633