Facilities must pursue the agile optimization of feedstocks and other inputs with products and operations to reflect market demand and prices. This is how the demand-pull business model is achieved and a measurable change in profitability delivered. This presentation will showcase why a mindset shift to value chain optimization is needed, as well as the deliberate approach needed to digitally transform value chain optimization activities. The value chain digital twin combining traditional solutions and AI will be profiled, along with the first steps that need to be taken, now.
8. Integration Model update Enterprise data
Current Time 1 to 15 days 1 to 3 months 1 year 2 to 10 years1 day1 month
IIoT data
Corporate business plans
Supply chain
planning
Variance analysis
Production
accounting
Strategic
Plan
Fiscal
Targets
Feed and
Inventory Plan
Operating Plan
Actual
Production
Investment
planning
Production
planning
Annual
planning
Supply chain
scheduling
Supply chain
management
Value chain optimizaton
Production
scheduling
Supply Chain
Status
LP
Least
Squares
Event-based simulation
Custom
Asset
Optimization
Process and off-sites automation Operating Targets
Actual
Operation
Schedule
Procedural automation APC
Model Update
Debottlenecking
Process and utility
optimization
Profit improvement
process unit management
Feasibility
study
Process
design
Process Simulation
Predictive
control
7
Integrated Systems and Practices
9. ■ Data driven, automated identification
of optimization using AI
■ Intelligent, automated work processes
■ Automated data management and
integration
■ Cloud enabled to facilitate:
■ Scalability
■ Collaboration
■ Support
■ Rapid enhancements
■ Knowledge management
■ Integration
■ Visualization
Digital Future
11. Data-driven Optimization
1
Knowledge graph for managing and
visualizing supply chain data
2 Demand, price and performance forecasting
3
Reconciliation, variance analysis and
opportunity identification
4 Automation of model updates
5 Automation of production scheduling
6 Asset-wide optimization
4
3
6
1
5
2
12. Value Chain Knowledge Graph
■ Global optimization
■ Visualize and manage entire
supply chain
■ Connect data silos–linked data
■ Add structure and context
■ ML, text mining and NLP
■ Semantic search and AI
■ Automation, classification
■ Reporting, personalization
Partner
2Partner
3
Plant 2
Plant 1
Place 2
Place 3
Place 1
Product
3
Product
1
Distanc
e
Cost
Time
Mode
Feed
Producti
on
Cost
Product
3
Production
database
Simulation database
Scheduling
database
Planning
database
ERP database
Unstructured
data
Business Taxonomy and Ontology—Meta Data
14. First principles
models
Actual
plant behavior
Data clean-up
and selection
Planning/scheduling models
Train
Calibrate
Update signal
Historical lab/plant data
On-line
monitoring
FCC Feed
Composition
Linearization
Point
Product
Yield
Linear
Model
Reality
Mismatch
Synthetic data
1) More variables
2) Less linear
3) Wider scope/more interactions
4) Invert model
5) Combine planning and scheduling
Automation of Model Updates
16. Integration with Petro-SIM provides
most accurate inferential
calculations and gain updates
Non-linear multi-unit or plant wide
optimization ensures operation
closer to global optimum
Dynamic optimization reduces time
to optimum operation
Tight integration with APC ensures
better disturbance rejection and
easier maintenance
Gain update assured by ML
application
On-line Petro-SIM rigorous model
Plant
Plant –wide APC and optimization
DCS
Sub Controller 1
Sub Controller 2
3
1/sec
1/5 mins
1/0.5~1hr
OPC
OPC
Assay or
feed properties
Economics
schedule
SS Gain Update
( If changed )
MV / CV Limit
ON / OFF Status
Historian
Plant-wide Optimization
27. The names of corporations, organizations, products and logos herein are either registered trademarks or
trademarks of Yokogawa Electric Corporation and their respective holders.
Thank You
Excellence
is never an accident. It is always the result of high
intention, sincere effort and intelligent execution; it
represents the wise choice of many alternatives—
choice, not chance, determines your destiny.