defining the business case for iot and machine learning in ... · •managed —measured; project,...
TRANSCRIPT
![Page 1: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/1.jpg)
AnalyticsXC o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Defining the Business Case for IoTand Machine Learning in Utilities
Alyssa FarrellGlobal Energy PracticeSAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
The promisehellip
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Todayrsquos realityhellip
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
The reality of IoT is here and utilities are already benefiting in quantifiable ways There are measurable steps you can take to achieve similar results
My premisehellip
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoSmart grid infrastructure is the first concrete example of the lsquoInternet of Thingsrsquo on a significant scale
Smart meters with two-way communications and other sensors across the power grid are already providing tangible benefits for utility customersrdquo
Ben Gardner President of Northeast Group
Metering
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
BETTER CUSTOMER SERVICE
ENABLING ENERGY EFFICIENCY PROGRAMS
IMPROVED DATA-DRIVEN DECISION MAKING
INCREASED DER INTEGRATION
BETTER CUSTOMER CHOICE ENGAGEMENT
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Customer interactions are online mobile social
Source Accenture The New Energy Consumer 2015
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquo[D]igital transformation is often not fully realized until a companyrsquos digital marketing strategy begins to inform and ultimately drive the larger business strategyrdquo
Digital Transformation Requires a Mindset Beyond MarketingDon Bulmer Gartner July 28 2016
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 2: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/2.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
The promisehellip
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Todayrsquos realityhellip
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
The reality of IoT is here and utilities are already benefiting in quantifiable ways There are measurable steps you can take to achieve similar results
My premisehellip
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoSmart grid infrastructure is the first concrete example of the lsquoInternet of Thingsrsquo on a significant scale
Smart meters with two-way communications and other sensors across the power grid are already providing tangible benefits for utility customersrdquo
Ben Gardner President of Northeast Group
Metering
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
BETTER CUSTOMER SERVICE
ENABLING ENERGY EFFICIENCY PROGRAMS
IMPROVED DATA-DRIVEN DECISION MAKING
INCREASED DER INTEGRATION
BETTER CUSTOMER CHOICE ENGAGEMENT
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Customer interactions are online mobile social
Source Accenture The New Energy Consumer 2015
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquo[D]igital transformation is often not fully realized until a companyrsquos digital marketing strategy begins to inform and ultimately drive the larger business strategyrdquo
Digital Transformation Requires a Mindset Beyond MarketingDon Bulmer Gartner July 28 2016
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 3: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/3.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Todayrsquos realityhellip
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
The reality of IoT is here and utilities are already benefiting in quantifiable ways There are measurable steps you can take to achieve similar results
My premisehellip
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoSmart grid infrastructure is the first concrete example of the lsquoInternet of Thingsrsquo on a significant scale
Smart meters with two-way communications and other sensors across the power grid are already providing tangible benefits for utility customersrdquo
Ben Gardner President of Northeast Group
Metering
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
BETTER CUSTOMER SERVICE
ENABLING ENERGY EFFICIENCY PROGRAMS
IMPROVED DATA-DRIVEN DECISION MAKING
INCREASED DER INTEGRATION
BETTER CUSTOMER CHOICE ENGAGEMENT
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Customer interactions are online mobile social
Source Accenture The New Energy Consumer 2015
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquo[D]igital transformation is often not fully realized until a companyrsquos digital marketing strategy begins to inform and ultimately drive the larger business strategyrdquo
Digital Transformation Requires a Mindset Beyond MarketingDon Bulmer Gartner July 28 2016
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 4: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/4.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
The reality of IoT is here and utilities are already benefiting in quantifiable ways There are measurable steps you can take to achieve similar results
My premisehellip
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoSmart grid infrastructure is the first concrete example of the lsquoInternet of Thingsrsquo on a significant scale
Smart meters with two-way communications and other sensors across the power grid are already providing tangible benefits for utility customersrdquo
Ben Gardner President of Northeast Group
Metering
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
BETTER CUSTOMER SERVICE
ENABLING ENERGY EFFICIENCY PROGRAMS
IMPROVED DATA-DRIVEN DECISION MAKING
INCREASED DER INTEGRATION
BETTER CUSTOMER CHOICE ENGAGEMENT
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Customer interactions are online mobile social
Source Accenture The New Energy Consumer 2015
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquo[D]igital transformation is often not fully realized until a companyrsquos digital marketing strategy begins to inform and ultimately drive the larger business strategyrdquo
Digital Transformation Requires a Mindset Beyond MarketingDon Bulmer Gartner July 28 2016
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 5: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/5.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoSmart grid infrastructure is the first concrete example of the lsquoInternet of Thingsrsquo on a significant scale
Smart meters with two-way communications and other sensors across the power grid are already providing tangible benefits for utility customersrdquo
Ben Gardner President of Northeast Group
Metering
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
BETTER CUSTOMER SERVICE
ENABLING ENERGY EFFICIENCY PROGRAMS
IMPROVED DATA-DRIVEN DECISION MAKING
INCREASED DER INTEGRATION
BETTER CUSTOMER CHOICE ENGAGEMENT
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Customer interactions are online mobile social
Source Accenture The New Energy Consumer 2015
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquo[D]igital transformation is often not fully realized until a companyrsquos digital marketing strategy begins to inform and ultimately drive the larger business strategyrdquo
Digital Transformation Requires a Mindset Beyond MarketingDon Bulmer Gartner July 28 2016
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 6: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/6.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoSmart grid infrastructure is the first concrete example of the lsquoInternet of Thingsrsquo on a significant scale
Smart meters with two-way communications and other sensors across the power grid are already providing tangible benefits for utility customersrdquo
Ben Gardner President of Northeast Group
Metering
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
BETTER CUSTOMER SERVICE
ENABLING ENERGY EFFICIENCY PROGRAMS
IMPROVED DATA-DRIVEN DECISION MAKING
INCREASED DER INTEGRATION
BETTER CUSTOMER CHOICE ENGAGEMENT
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Customer interactions are online mobile social
Source Accenture The New Energy Consumer 2015
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquo[D]igital transformation is often not fully realized until a companyrsquos digital marketing strategy begins to inform and ultimately drive the larger business strategyrdquo
Digital Transformation Requires a Mindset Beyond MarketingDon Bulmer Gartner July 28 2016
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 7: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/7.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
IoT investments today (amp tomorrow)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
BETTER CUSTOMER SERVICE
ENABLING ENERGY EFFICIENCY PROGRAMS
IMPROVED DATA-DRIVEN DECISION MAKING
INCREASED DER INTEGRATION
BETTER CUSTOMER CHOICE ENGAGEMENT
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Customer interactions are online mobile social
Source Accenture The New Energy Consumer 2015
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquo[D]igital transformation is often not fully realized until a companyrsquos digital marketing strategy begins to inform and ultimately drive the larger business strategyrdquo
Digital Transformation Requires a Mindset Beyond MarketingDon Bulmer Gartner July 28 2016
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 8: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/8.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
BETTER CUSTOMER SERVICE
ENABLING ENERGY EFFICIENCY PROGRAMS
IMPROVED DATA-DRIVEN DECISION MAKING
INCREASED DER INTEGRATION
BETTER CUSTOMER CHOICE ENGAGEMENT
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Customer interactions are online mobile social
Source Accenture The New Energy Consumer 2015
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquo[D]igital transformation is often not fully realized until a companyrsquos digital marketing strategy begins to inform and ultimately drive the larger business strategyrdquo
Digital Transformation Requires a Mindset Beyond MarketingDon Bulmer Gartner July 28 2016
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 9: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/9.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Customer interactions are online mobile social
Source Accenture The New Energy Consumer 2015
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquo[D]igital transformation is often not fully realized until a companyrsquos digital marketing strategy begins to inform and ultimately drive the larger business strategyrdquo
Digital Transformation Requires a Mindset Beyond MarketingDon Bulmer Gartner July 28 2016
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 10: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/10.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquo[D]igital transformation is often not fully realized until a companyrsquos digital marketing strategy begins to inform and ultimately drive the larger business strategyrdquo
Digital Transformation Requires a Mindset Beyond MarketingDon Bulmer Gartner July 28 2016
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 11: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/11.jpg)
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
Crossing the divide
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 12: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/12.jpg)
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics Capabilities to cross the divide
Data
DeploymentDiscoveryTo get VALUE out of any analytics endeavor 3 key components have to be considered
bull DATA
bull DISCOVERY and
bull DEPLOYMENT
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 13: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/13.jpg)
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Data is the foundation
Structured Data
Onl ine Digital Data
Machine Data
Social Media Data
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 14: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/14.jpg)
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
What we do with Data DISCOVERY
Visualization
Prediction
Machine Learning
Optimization
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 15: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/15.jpg)
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
How we Operationalize the results DEPLOYMENT
Data Warehouse
CRM Cal l Center
Mobile Channel
Devices
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 16: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/16.jpg)
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Big data analytics maturitybull Ad hoc mdash experimental siloed proof-of-concept or pilot projects undefined
processes lack of resources and individual efforts
bull Opportunistic mdash accepted recurring projects budgeted and funded program management and documented strategy and processes with stakeholder buy-in
bull Repeatable mdash intentional defined requirements and processes unbudgeted funding and project management and resource allocation inefficiency
bull Managed mdash measured project process and program performance measurement influences investment decisions and standards emerge
bull Optimized mdash operationalized continuous and coordinated BDA process improvement value realization
IDC MaturityScape Benchmark Big Data and Analytics in European Utilities
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 17: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/17.jpg)
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Value of Analytics
5x
Year-over-year
increase in
positive
social media
sentiment
22x
Annual
improvement in
net promoter
scores
More likely to
optimize
productservice
bundles for
consumers
384x
Year-over-year
increase in
customer program
enrollments
(all industries)
Source How Analytics Reveals New Util ity Customer Value Aberdeen Research amp SAS
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 18: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/18.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ldquoRetaining satisfied customers in deregulated energy markets requires a combination of the best energy services the right customer communication channels and optimal marketing campaignsrdquo
Yetik Mert
CEO Enerjisa
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 19: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/19.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 20: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/20.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine Learning Internet of Things
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 21: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/21.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 22: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/22.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
History As early as 1950s Alan Turing and Arthur Samuel define the
concept of machine learning as the ability of computers to learn to function in ways that they were not specifically programmed to do
1950s 1980s 2000s
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 23: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/23.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Machine learning ndash fact or fiction10 Ways Machine Learning Is Revolutionizing Manufacturing
Forbescom Jun 26 2016
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 24: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/24.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
ML Algorithms Neural networks
Decision trees
Random forests
Associations and sequence discovery
Gradient boosting and bagging
Support vector machines
Nearest-neighbor mapping
k-means clustering
Self-organizing maps
Regression
Expectation maximization
Multivariate adaptive regression splines
Bayesian networks
Kernel density estimation
Principal components analysis
Singular value decomposition
Model ensembles
Local search optimization techniques such as genetic algorithms
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 25: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/25.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 26: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/26.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 27: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/27.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 28: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/28.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 29: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/29.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Cybersecurity by the numbers
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 30: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/30.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
THE SMART GRID ANALYTICS OPPORTUNITIES ABOUND
Optimize
Products and Services
Predict
Transformer Failure
Forecast
Future Power
Demands
Bring Big
Data to the Desktop
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 31: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/31.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 32: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/32.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Smart grid analytics
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 33: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/33.jpg)
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Conclusion
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 34: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/34.jpg)
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Deploy
IoT Analytics Lifecycle
ETL
Data Data Storage
Alerts Reports Decisioning
De
plo
y
f
Data Streams Intelligent Filter Transform
Streaming Model Execution
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 35: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/35.jpg)
analyticsx
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
Tips for embracing IoT and ML1 Develop a strategy
2 Pick a project
3 Donrsquot overlook change management
4 Learn from other industries
5 Plug into other utilities
6 Invest wisely
6 Steps to Becoming a Utilities loT Ninja
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX
![Page 36: Defining the Business Case for IoT and Machine Learning in ... · •Managed —measured; project, process, and program performance measurement influences investment decisions and](https://reader030.vdocuments.us/reader030/viewer/2022041021/5ed1b4077dccd150e82adbaf/html5/thumbnails/36.jpg)
C o p y r ig ht copy 201 6 SAS In st i tute In c A l l r ig hts r ese rve d
AnalyticsX