2. 2
Global Learnings of Digital Agenda for Refinery 4.0
World Economic
Council Energy Monitor
2018 Report
As the innovation
cluster rises in the
energy leaders’ agenda,
there is a key focus on
issues linked to
digitalization and the
action needed to
facilitate the
convergence of new
energy technology
including Data AI,
Mobile Cloud and Block
chain.
3. 3
Global Learnings of Digital Agenda for Refinery 4.0
Digital Innovation Agenda for Global Downstream Companies
with TCS as Partner
Strong R & D, Engineering & Domain Practice, Data Science
Leadership in Digital Re-imagination and Business 4.0/ Industry 4.0
Strong Focus on Agile & Design Thinking Both in Solution Conceptualization & Execution
Research Community Support & Startup Eco System
Cross Industry excellence in delivery globally
TCS recent engagement - : The Objective of Refinery 4.0 Centre of Digital Innovation is:
Develop explore, Conceptualize, Architect & deploy Digital & Game Changing technologies
across Total Business value chain that can provide a huge value and redefine their business
Visit: www.tcs.com/energy-resources-utilities
4. 4
HORIZON 1 HORIZON 2 HORIZON 3
TECHNOLLOGY
TYPE
(As
seen
from
Market)
TIME HORIZON (Time to deliver MVP)
Natural language
processing
Machine learning
Internet of
Things
AR/VR
Image analysis
Digital Replica-
Status
Data fusion
Deep Learning
Multispectral
Image analysis
EMERGING
PILOT
KEY
TCS research & development foot prints in Digital
What are technology types?
Emerging technologies are 3 + years
away from majority adoption.
They have some uncertainty but are very
promising
Pilot technologies are 1 - 3 years away
from majority adoption.
They are being piloted today and are
gathering momentum.
Mature technologies are in market and
are being adopted for many applications.
AI Neural Automation
What is Time Horizon?
HORIZON 1
Time to deliver MVP at pilot site
In 12 months
HORIZON 2
Time to deliver MVP at pilot site
In 12-24 months
HORIZON 3
Time to deliver MVP at pilot site
In 24+ months
In 12 months In 12-24 months In 24+ months
Drones
Digital Replica-
Predictive
Blockchain
Cloud
computing
Robotic Process
Automation
AI
5. 5
Opportunities
Business dynamics
• Improve Long Term Profitability
• Beat Volatile Market
• Help Prevent Climate Changes &
• Reduce Carbon Footprints of Process & Products
Operation Intelligence
• Embed Continuous Improvement
Future Refinery
• Optimize Costs & Availability,
• Optimize Energy Efficiency,
• Be Agile and Flexible,
• Transform the Way You Operate,
Inspect,
Maintain, Supervise & Optimize
🗠
🏭
🛢
🛢
📊
Thorough Analytical Capabilities & Automation
in Decision Making & Information Management
Multiple Challenges
⛓️ Value Chain
🏭 Changes to the Production and Service Mode
⚽ Integrated Management & Control Chain
6. 6
Leveraging Industry 4.0
Data Becomes Key in Decision Making for Process Management &
Optimization
Enable Agility & Integration
Equipment level
Systems level
Production units
Availability as the Main Driver
for Performance of the Refinery
🛢
🏆
🏎
The integration of optimization features at levels -
from equipment to plant is an important function of
the production management system
Click to
Read Post
7. 7
Defining Refinery 4.0
Digital
Mission
for
Refining
Smart
Operation
Management
Smart
Assets,
Digital Twins
& Smart
Staff
Digital
Blending
& Predictive
Product
Quality
Automate
Compliance &
Regulatory
Management
Enhanced
Personal Safety
with Wearables
&
Cyber Security
Integrated
Supply Chain
& Global
Trade
automation
Digital Silos Connected refinery Predictive Refinery Adaptive Refinery
Maturity Journey
Business Processes Business Models Reimagine Work
Reimagine Downstream
Refinery 4.0 will be lean.
It will seamlessly integrate all the
principles of lean manufacturing
using digital technologies with plug-
n-play standardization enhancing
agility
Leverage Industry 4.0
Interconnectivity
Information Transparency
Augmented Decisions
Decentralized Decisions
9. 9
Data Engineering for Plant Analytics
Numerous data clusters
Islands of information
Source: different systems at different locations
Individualistic and lack comprehensive information of the entire system
The same data may be available at multiple locations (i.e., duplicity of data)
Data is generated from disparate systems and applications across the
organization
Raw data from edge device,
instrumentation and control
systems, data acquisition
systems, process control
etc.
CRM, ERP,
Finance, Sales,
etc.
Governance
information, HSE,
planning and
scheduling, asset data,
customer service, etc.
Processed data
from production
systems, LIMS,
Energy, yield
accounting, etc.
Process
Data
Business
Data
Master
Data
Production
Data
Aggregate, Contextualize,
Analyze, Visualize
Data location
Data speed
Data type
Analytics model
Statistical, first principles, artificial
intelligence (AI), machine learning (ML)
Purpose
Display, long-term improvement, direct
feedback to control system, improve
people processes
“Intelligizing” the data
10. 10
Smart Asset : Digital Twin
Use
Cases
Use analytics to predict asset failure to reduce
outages and inspections.
Automated notifications of equipment
performance
Ability to integrate to other processes such as
scheduling service
Assets monitor themselves and their peers and
react to reroute, shed load, or shut down to
minimize outage and asset damage.
Real-time awareness of the asset condition
through dense deployment of wireless and wired
sensors.
Augmented and virtual reality are used to provide
technicians with relevant information and guided
work instructions.
…. Refer IDC Report
Manufacturing Process Digital Twin Model
Physical Asset
Fleet
Aggregate
Data
Operational
History
Maintenance
History
Real Time
Operational Data
Digital
Twin
Physics Based Models
Statistical Models
Machine Learning
FMEA
CAD Model
FEA Model
11. 11
The Data Intelligence building for Digital Twin in downstream
Manufacturing Execution
System
Supervisory Control & Data Acquisition / Distributed
Control System
Sensors & Actuators
Plant
Measurements Signals
Set Points
Measurement
s
Advanced Process Optimization &
Control
Measure
Enterprise Resource
Planning
Market Demand Supply Chain
Laboratories
Laboratory
Information
Management
System
@ TCS Proprietary and Confidential
ANALYZE
Process Models
SOFT-
SENSE
12. 12
the #digital-replica consists of three components
Data Model
A systematic way to represent the variety
of data related to the assets and
processes.
Analytics
Algorithms that Describe, Predict and Prescribe the
behaviour of an asset or process.
(a) Thermodynamics + Chemistry + Physics …
(b) Data Driven - AI + Natural Language Processing
+ …
(c) …
Knowledge base
Data sources that feed analytics, CUSTOMER
Expertise, subject-matter expertise, historical
data, industry best practices …
The Digital Replica
13. 13
Predictive analytics using IIoT and machine learning for detection and
prediction of failure
IIOT Integration with Existing OT Data Fabric & Analytic
IIOT
Data
DCS
SCADA
Yield
Accenting P&S
Unit Models
Financial Data
ERP
Laboratory Data
LIMS
Integrated Refinery
Data
Equipment
Data
EAM
PI Systems
OT Object Model
Opralog
OT Data Mode/
Infrastructure
E-Logbook
PI Integrator
for Azur
IIoT capturing data of thousands of sensors,
for performance monitoring &
control like 100000 data points
Use of Microsoft Azure Machine Learning for
proactive development of several hypotheses
on how to improve the coking process and
reduce the risk of steam eruptions
Leveraging existing data in OSIsoft’s PI
System, for better to analyze operational data
quickly without needing to invest in a whole
new system.
Reference : Global Refinery case study published on :
news.microsoft.com/europe/2017/05/05/refining-oil-in-the-cloud/
1
2
3
14. 14
Tool Kits for Refinery 4.0
Introducing TCS R&D digital
advance tool sets
Digital Twin Framework : PEACKOCK
iSense - Actionable Insights from Internet of Things
Based on research in deep learning
Premap : A Digital Enabling Platform for Knowledge, Data
and Simulation driven Integrated Engineering Analysis
Using Text mining, AI
Discussion use case applicability of these tools
17. 17
Key take ways
All global Oil majors has digital transformation agenda for refineries
Identification of focus areas for digitalization at all levels and phase wise
approach
Policy for data quality and Data science for downstream
Digital being new for downstream, Innovation as strategy
Analytics as a service line
Digital at Organization level
Learning from Other industries and experiences of vendors
18. 18
TCS Focus on Business 4.0
Tailor/Mass
Customize
ABUNDANCE
Framework
Leverage
Ecosystems
Embrace Risk
Create
Exponential Value
INTELLIGENT
AGILE
AUTOMATED
CLOUD
19. 19
Energy & Resources Unit Snapshot
5 % of TCS revenues
15000 + Experienced Consultants
70+ Customers - Footprint across the globe
Americas, Europe, APAC, Africa & Middle East
Key Academic relationships: MIT, Rice University, Texas A&M
Analyst Recognition: Industry Leaders in Energy by IDC
and WINNERS in Energy Operations Services by HfS
Research
Broad range of services across ERU segments: Oil & Gas, Oil
Field Services, Alternative Energy, Metals, Mining, EPC
and Utilities
Successfully Developed
Industry Specific Solution and 20+
Horizontal Solutions
Provide Consulting, Outsourcing,
Engineering & Industrial,
Digital services to our Customers
Footprint across the globe
Americas, Europe, APAC,
Africa & Middle East
Trusted
Research & Innovation
Partner
20. 20
&
M D Agrawal
Advisor & Director downstream COE, TCS
Global Oil & Gas practice
agrawal.murli@tcs.com
https://www.linkedin.com/in/m-d-agrawal-7a792510/