IoT & Data Analytics Sharing Session - Telkomsigma

© PT. Sigma Cipta Caraka 2019
IoT & Data Analytics
TOGI NABABAN - Telkomsigma
2019

2
IoT & Big Data Analytics
about
IoT & Data
Analytics
Collecting Information from lots of devices
is cool – but it’s just telematics Merging
perspectives between devices, systems and
humans to build a better understanding of the
world around us..
Then tying together insightwith action – there
lies the promise of IoT

3
IoT & Big Data Analytics
Internet of Things = “is a computing
concept that describes the idea of everyday
physical objects being connected to the
internet and being able to identify themselves
to other devices.”
IoT Analytics requires Big Data Analytics, which
are include Streaming Data Management (Data
In-motion Analytics at Edge) and Big Data
Analytics (Data At-rest Analytics at Cloud)
Big Data Analytics = “Massive Information in the
scale of Volume, Velocity, Variety,
Variability and Value”

4
MarketReferences
Market
References

5
MarketReferences:UnderstandingIoT Drivers
Chinese enterprises are most
often adopting IoT to increase
competitiveness (23%) while
American companies are
focused on reducing costs (19%)

6
WHY BuildingIoT Solution

7
Industry 4.0 Opportunity
FurtherElaboration:IoT For Industry
From Isolated, Optimized Cell…
… to fully integrated data &
product flows across border
Fully Integrated Cyber-Physical
Source: Mckinsey, 2016

8
Industry4.0:Revisited
Enhancing mechatronic
components with embedded
systems
Linked physical object
We’re getting there..
Next slide
Expose services..

9
MakingIndonesia4.0

10
Industry4.0:ImpactOn Manufacturing
Source: Rolland Berger, 2015

11
Industry4.0:Data & CommunicationAs Backbone
Source: Rolland Berger, 2015

12
Industry4.0:SmartFactory
Source: IoT Analytics

13
WHY Industry4.0
At present, when an event occurs in a company there is a delay
before detailed insights about the event become available. This
means that there is also a delay in taking the corresponding
decisions and (counter-)measures.
Industry 4.0 capabilities help manufacturing companies to
dramatically reduce the time between an event occurring and
the implementation of an appropriate response.
Source: acatech STUDY

14
Stage-1
Computerization
Stage-2:
Connectivity
Stage-3:
Visibility
Stage-4:
Transparancy
Stage-5:
Predictive
Capacity
Stage-6:
Adaptability
Industry4.0:DevelopmentPath
Descriptive Diagnostic Predictive Prescriptive
ANALYTIC CONTINUUM, by Gartner
Source: acatech STUDY

15
Industry4.0:DevelopmentPath
Stage-3:
Visibility
Stage-2:
Connectivity
Stage-1:
Computerization
Stage-5:
Predictive
Stage-4:
Transparency
Stage-6:
Adaptability
❑ AS-IS: Isolated deployment of IT&OT.
❑ Connectivity between all process.
❑ Implementation: IoT Gateways, Networks
❑ Why Something is Happening
❑ Aggregates of Data, Complex Event
Processing
❑ Implementation: Big Data Analytics, BI
Dashboard
❑ Automated Decision Making & Action
❑ Robotics
❑ AS-IS: Machines without digital interface,
manual operations
❑ Device Digitalisation
❑ Implementation: Digital Sensor
❑ What is Happening
❑ Digital Shadow (Up-to-date digital model of
factory); Data integration (PLM, ERP, MES)
❑ Real-time Monitoring (Centralize Data
Management)
❑ Daily Operation Reports
❑ Implementation: Data Management,
Monitoring Dashboard
❑ Prediction
❑ Advance Big Data Analytics
❑ Predictive Analytics
❑ Implementation: Artificial Intelligence
(Machine Learning)

16
ValueChain:IoT & Data Analytics
DEVICE
PROVIDER
NETWORK
PROVIDER
APPLICATION
PROVIDER
PLATFORM
PROVIDER

17
IoT ValueChain:Device(s)
DEVICE PROVIDER

18
IoT ValueChain:Connectivity
TYPE OF
CONNECTIVITY

19
IoT ValueChain:NeworkService/DeviceProvider
NETWORK
(SERVICE/DEVICE)
PROVIDER
NETWORK SERVICE PROVIDER
❑ Network Service Provider:
Provide bandwidth or network
access by providing direct Internet
backbone access to internet service
providers.
❑ Network Protocols:
Sessions: Server-2-Server(S2S),
Device-2-Server (D2S), Device-2-
Device (D2D).
Data Link: Short Range, Long
Range, Tethered.
NETWORK SERVICE PROVIDER

20
IoT ValueChain:PlatformProvider
PLATFORM
PROVIDER
❑ Open Source IoT & Data Analytics Platform
Gartner, Industrial IoT Platforms, Feb 2018
Gartner, Analytics & BI Platforms, Jan 2019
❑ Enterprise IoT & Data Analytics Platform

21
IoT ValueChain:PlatformProvider
PLATFORM
PROVIDER
Device Gateway Rules Engine Message
Broker
Device
Shadow
Device
Registry
DATA MANAGEMENT DEVICE MANAGEMENT
IoT Pillars

22
ExampleProductRoadmap:Cloud-Ready
OBJECTIVES
❑ Owned Product IoT Platform (TELKOM
Group)
❑ To Build Cloud-Ready Deployment IoT
Platform as Cloud Content (PaaS)

23
IoT ValueChain: VerticalSolutions
APPLICATION PROVIDER
(VERTICAL SOLUTIONS)

24
IoT Ecosystem:VerticalSolutions
APPLICATION PROVIDER
(VERTICAL SOLUTIONS)
PERSONAL
VEHICLES
ENTERPRISE INDUSTRIAL

25
IoT Use Case:SMART FLEET
USE CASE 01
SMART FLEET
USE CASE 02
SMART ENERGY
USE CASE 03
SMART METERING
USE CASE 04
SMART FARMING
OBJECTIVES:
❑ Reliable and fault-tolerant data
collection from your IoT devices and
sensors to monitor facilities state, crop
growth characteristics, humidity level,
etc.;
❑ Advanced and flexible data visualization
for real-time and historical monitoring
of future farms;
❑ Customizable end-user dashboards to
share farm monitoring results;
❑ Integration with third-party analytics
frameworks and solutions for advanced
analytics and machine learning;
❑ Optimize returns on inputs while
preserving resources by remotely
configuring IoT devices based on results
of the analytics.

26
IoT Use Case:SMART ENERGY
USE CASE 01
SMART FLEET
USE CASE 02
SMART ENERGY
USE CASE 03
SMART METERING
USE CASE 04
SMART FARMING
OBJECTIVES:
❑ Reliable and fault tolerant data
collection for your smart meters and
energy monitors;
for real-time and historical smart
energy monitoring;
analyse and share the results of energy
efficiency monitoring;
electricity usage monitoring;
❑ Enable energy management by utilizing
API to control and manage smart
meters.

27
IoT Use Case:SMART METERING
USE CASE 01
SMART FLEET
USE CASE 02
SMART ENERGY
USE CASE 03
SMART METERING
USE CASE 04
SMART FARMING
OBJECTIVES:
❑ Reliable and fault tolerant data
collection for smart water meters,
energy monitors, smart energy meters,
etc.
❑ Advanced, customizable data
visualization for real-time and historical
smart metering monitoring.
❑ Alarm widgets to instantly notify users
and / or operators about any critical
events or unusual consumption levels.
❑ Device management to allow to
organize endpoints in groups by
specific attributes.
processing of smart metering data and
reporting.

28
IoT Use Case:SMART FARMING
USE CASE 01
SMART FLEET
USE CASE 02
SMART ENERGY
USE CASE 03
SMART METERING
USE CASE 04
SMART FARMING
OBJECTIVES:
❑ Reliable and fault-tolerant data
collection from your IoT devices and
sensors to monitor facilities state, crop
growth characteristics, humidity level,
etc.;
for real-time and historical monitoring
of future farms;
share farm monitoring results;
analytics and machine learning;
❑ Optimize returns on inputs while
preserving resources by remotely
configuring IoT devices based on results
of the analytics.

29
IoT For SmartRailwaysSystem
Stage-1
•Sensor
Stage-2
•Connect
Stage-3
•IoT
DATA
TO CLOUD
DATA
TO CLOUD
[Train Sensors] INTRA-TRAIN:
1. HVAC Sensor (Heating, Ventilation and Air
Conditioning).
2. Engine Temperature.
3. Electrical Generator & Voltage.
4. Water Tanks.
5. Battery Charge Monitoring.
6. Compartment Control.
7. Speed Measurement.
8. Lateral Vibration.
9. Brakes & Tractions.
Pilot
[Wearbles Device]:
1. Activity Type Monitoring.
2. Number of Steps.
3. Distance Travelled.
4. Vital Sign Reading (Pulse Rate Sensor,
Body Temperature Sensors).
5. Motion Data.
[Train Sensors] THE ENGINE:
1. Temperatures: Engines, Radiator, Motor, Bogey.
2. Lateral Vibration.
3. Battery Voltage & Charge Monitoring.
4. Liquid Bar Pressure.
5. Voltage: Altenator, Rectifier, Inverter.
6. Fuel Tank Meter.
7. Air Compressor.
[Train Sensors] INTER-TRAIN:
1. Coupler Carrier Plate & Cross Key.
(Coupler securement, missing fastener).
2. Spring & Wedge
3. Undercarriage (Frame inspection)
4. Breaks Health.
5. Wheel Profiles (wear limits).
6. Inter-Car air hose height.
[TRAIN SENSOR]
SERVER
STORAGE
IoT Gateway
DATA
ON STAGING
DATA
ON STAGING
[TRAIN SENSOR]
IoT Gateway
IoT Platform
CLOUD COMPUTING
EDGE COMPUTING

30
IoT For SmartRailwaysSystem
Stage-4
•Data
Analytics
Stage-5
•Predictive
Maintenance

31
about
Data Analytics

32
Skill Set
Team:Skills,Roles& Responsibility
Roles &
Responsibility
 Collect Data → AnalyzeData →
Build Report.
 Data Understanding.
 Data Acquisition & Maintenance
 Data Cleansing & Integration
 Statistical Analyses & Data
Interpretation
 Pattern Identification & Analysis
 Reporting & Data Visualization
 OptimizeStatistical Efficiency &
Quality
 Spread-sheet& SQL/Database
Knowledge
 Data Warehousing
 Scripting & Statistical Knowledge
 Programming Knowledge
(Phyton/R/SAS)
 Reporting & Data Visualization
Data Analyst & Visualization
 Setting-up Data Pipeline.
 Develop, Construct, Tests and
Maintains The Complete
ArchitectureOf Large Scale
Processing System
 Develop, Test & Maintain
Architecture
 Develop DatasetProcess
 Deploy Analytics, Statistical &
Machine Learning Platform
 Predictive & PrescriptiveModelling
 Find Hidden Pattern
 Data Architecture
 Data Warehousing & ETL
 In-depth KnowledgeSQL/Database
 Hadoop-based Analytics
 Advanced Programming Knowledge
(Phyton/R/SAS)
 Machine Learning Concept
Knowledge
Data Engineer
 Visualization & Business Decision
Making
 ProfessionalComplexData Analytics
With Expertise in Scientific
Disciplines
 Data Mining
 Develop OperationalModels
 In-depth Machine Learning
Optimization
 Data Enhancement & Sourcing
 Strategic Planning For Data Analytics
 Ad-hoc Analyses& Anomaly
Detection
 Statistical & Analytical Skill
 Data Mining Knowledge
 Machine Learning & Deep Learning
Principles
 In-depth Programming Knowledge
(Phyton/R/SAS)
Data Scientist

33
AnalyticsFramework:Data Mining
❑ Find the right models
❑ There is no single Solution fit all – Need
to find the right approach, with the right
objectives
❑ To Build Use Cases
CROSS INDUSTRY STANDARD PROCESS FOR DATA MINING

34
Algorithm Taxonomy:To BuildAnalyticModel
Source: IIC Analytic Frameworks

35
Data AnalyticRoadmapForBankingSolution
OBJECTIVES
❑ Handle Big Data (3V: Volume, Variety, Velocity)
❑ Integrate Multiple Data Source (Silo: CORE Banking, Digital Services, Other
Data)
❑ Reduce Cost (ETL Process, Analytical Process, Silo Data Platform, On-Cloud
Platform Feasibility)
❑ Enhance Capability (Unstructured Data Analytics, Advance Analytics)
❑ Reduce Time To Market (Faster Data Processing/Analytics)

36
GeneralArchitecture:Data AnalyticsForBanking
DATALAKE
INTERNET BANKING
MOBILE BANKING
AGENT BANKING
SMART BRANCH
EDC & ATM
CORE BANKING CONVENTIONAL
CORE BANKING SATU
CORE BANKING SHARIA
ETL/CRAWLER/
DATA ACCESS
DOMAIN
• CUSTOMER
• SUPPLIER
• PRODUCT
• EMPLOYEE
• ASSET
• DATA PROFILE
• DATA 360
VISIBILITY & ACCOUNTABILITY
• CROSS BU
• CROSS FUNCTIONAL
• CROSS DEPARTMENT
DATA WAREHOUSE
(OLAP)
DESCRIPTIVE ANALYTICS
DIAGNOSTIC ANALYTICS
PREDICTIVE ANALYTICS
PRESCRIPTIVE ANALYTICS
HADOOP
SOCIAL NETWORK
ANALYTICS; TEXT ANALYTICS
EXECUTIVE
DASHBOARD
OPERATIONAL
DASHBOARD
REPORTING REPO
RULE ENGINE
RULE-SET-01
RULE-SET-02
RULE-SET-n
EXCEPTION IDENTIFICATION
(FRAUD ANALYTIC)
REAL TIME DB
FRAUD RULE-SET
CAMPAIGN MANAGEMENT
SYSTEM
STANDARD
REPORTS
CASE MANAGEMENT
(FOR FRAUD MITIGATION)
NOTIFICATION SYSTEM

37
BankingSolution:UseCase 01
USE CASE 01
Customer
Profitability Analysis
OBJECTIVES:
❑ Menyediakan Tools Yang Dapat
Menunjukkan Profitability Detail Dari
Setiap Customer Untuk Memberikan
Penawaran Yang Bertarget Dengan
Produk Yang Tepat.
❑ Menawarkan Layanan Perbankan Yang
Sesuai Dengan Nasabah.
TARGET VALUES:
❑ Menyediakan Informasi Real-time
Untuk Diakses Oleh Customer Service
(Front Liner).
❑ Memberikan Rekomendasi List
Nasabah Yang Berpotensi Untuk
Dilakukan Penawaran Berdasarkan
Pengelompokan Tertentu.
USE CASE 02
CHURN
Analysis
USE CASE 03
BEHAVIOR
Score
USE CASE 04
CREDIT
Score Risk
USE CASE 05
CUSTOMER
Segmentation
USE CASE 06
PRODUCT
Recommendation

38
USE CASE 01
Customer
USE CASE 02
CHURN
Analysis
OBJECTIVES:
❑ Menyediakan Tools Yang Dapat
Menunjukkan List Nasabah Yang
Memiliki Tingkat Kencenderungan
Untuk Pindah Menggunakan Produk
Kompetitor (Churn).
TARGET VALUES:
❑ Menyediakan Informasi Real-time
Untuk Diakses Oleh Customer Service
(Front Liner).
❑ Memberikan Rekomendasi List
Nasabah Yang Berpotensi Pindah
(Churn).
USE CASE 03
BEHAVIOR
Score
USE CASE 04
CREDIT
Score Risk
USE CASE 05
CUSTOMER
Segmentation
USE CASE 06
PRODUCT
Recommendation

39
BankingSolution:UseCase-3
USE CASE 01
Customer
USE CASE 02
CHURN
Analysis
USE CASE 03
BEHAVIOR
Score
BENEFITS:
❑ Segmenting Customers: Providing
Recommendations About High-Risk,
Medium Or Low-Risk Customers To Be
Offered Supplementation.
❑ Personal Treatment: Determining
Campaigns Or Caring Programs Based
On Customer Scoring Or
Segmentation.
❑ Effective Resource: Increasing
Effectiveness And Efficiency In Terms
Of Time, Money And Other Resources
❑ Algorithm: Weight Of Evidence (WOE)
And Information Value (IV) Are Simple,
Yet Powerful Techniques To Perform
Variable Transformation And Selection
USE CASE 04
CREDIT
Score Risk
USE CASE 05
CUSTOMER
Segmentation
USE CASE 06
PRODUCT
Recommendation

40
USE CASE 01
Customer
USE CASE 02
CHURN
Analysis
USE CASE 03
BEHAVIOR
Score
USE CASE 04
CREDIT
Score Risk
BENEFITS:
❑ Memprediksi Performansi
Pengembalian Kredit Pada Pemohon
Pinjaman Untuk Mencegah
Bertambahnya Resiko Gagal Bayar /
Non-PerformingLoan (NPL)
USE CASE 05
CUSTOMER
Segmentation
USE CASE 06
PRODUCT
Recommendation

41
BankingSolution:Usecase-5
USECASE-1
Customer
USECASE-2
CHURN
Analysis
USECASE-3
BEHAVIOR
Score
USECASE-4
CREDIT
Score Risk
USECASE-5
Customer
Segmentation
BENEFITS:
❑ Customer Lifetime Value Enables Your
Business To Classify Different
Customer Groups And Different
Potential Customer Groups By Long
Term Profitability.
❑ Two Fundamental Tactics In Any
MarketingProgram Are To Up-Sell And
Cross-Sell.However,Which One Is The
Best Option? When To Choose And On
What Segment?. Customer Lifetime
Value Could Give You A Guideline To
Make A Decision And Investment On
Up-SellAnd Cross-Sell.
❑ Customer Segmentation: Generator,
Passer,Leaker,Saver
Sumber: Tim Data Scientist Telkom DDS
USECASE-6
PRODUCT
Recommendation

42
BankingSolution:Usecase-6
USECASE-1
CUSTOMER
USECASE-2
CHURN
Analysis
USECASE-3
BEHAVIOR
Score
USECASE-4
CREDIT
Score Risk
USECASE-5
CUSTOMER
Segmentation
USECASE-6
PRODUCT
Recommendation
BENEFITS:
❑ Analysis and segmentationof
transaction data and customer profiles
of product variants owned by the
client.
❑ Analysis result to be used for:
❑ Product Cross Sales: to offer
related product.
❑ Product Up Sales: to offer a higher
product spec.
❑ Both: Product Cross-Up Sales.
Sumber: Tim Data Scientist Telkom DDS
TARGET VALUES:
❑ Different benefit offers can be done
together more effectively.
❑ Can add customer fee-based income.
❑ Customers can obtain other products
according to their needs.

DEMO
IOT SIMULATION
WITH
MICROSERVICES

45
FleetTracking– IoT Mock-UpOn Microservices
Components
MQ
Position
Simulator
PositionTracker
WebAPP
MongoDB
WHY Microservices
❖ LOOSELY Coupled
❖ HIGHLY Cohesive

46
WHENTo Use Microservices
Microservices Provide BENEFITS…
•Strong Module Boundaries:
Microservices reinforce modular structure,
which is particularly important for larger
teams.
•Independent Deployment:
Simple services are easier to deploy, and
since they are autonomous, are less likely
to cause system failures when they go
wrong.
•Technology Diversity:
With microservices you can mix multiple
languages, development frameworks and
data-storage technologies.
… but come with COSTS..
•Distribution:
Distributed systems are harder to program,
since remote calls are slow and are always
at risk of failure.
•Eventual Consistency:
Maintaining strong consistency is
extremely difficult for a distributed system,
which means everyone has to manage
eventual consistency.
•Operational Complexity:
You need a mature operations team to
manage lots of services, which are being
redeployed regularly.

IoT & Data Analytics Sharing Session - Telkomsigma

Recommended

Recommended

More Related Content

Similar to IoT & Data Analytics Sharing Session - Telkomsigma

Similar to IoT & Data Analytics Sharing Session - Telkomsigma (20)

Recently uploaded

Recently uploaded (20)

IoT & Data Analytics Sharing Session - Telkomsigma