Downstream Logistics Optimization
Downstream and midstream energy facilities include a complex network of personnel, equipment, processes, and infrastructure. Maintaining base operations involves safe, reliable, and efficient movement of hydrocarbons into and out of sites to support the manufacturing and delivery of products to market.
The management of logistics operations is largely handled with paper-based and manually-intensive workflows, exposing companies to increased financial and field risk. Hydrocarbon distribution networks are diverse, and companies must effectively monitor, analyze, and optimize continuous operations.
The advent of energy-focused cloud technology now gives the industry the capability to improve the visibility and management of operations – orchestrating critical processes, automating workflows, and allowing cross-functional personnel to work efficiently together. AWS’ Downstream Logistics Optimization is a cloud-native solution to improve the tracking, analysis, forecasting, and optimization of hydrocarbon logistics operations. By combining geospatial intelligence and machine-learning services, the solution improves how schedulers, operators, and field personnel manage logistics operations of feeds, intermediates, and products - to improve business costs, field efficiency, utilization, and lower risk.
Solution Components
Track
Monitor
Monitor for anomalies and notify relevant personnel.
Forecast
ETA predictions based on real-time variables.
Optimize
Enable operational changes for field efficiency and incremental value.
Integrate
Integrate with business applications and Contact Center.
Value Drivers
Higher Utilization
Improved Agility
Lower Operating Cost
Incremental Margin Profit
Lower Field Risk
Customer Case Study
Challenge:
TC Energy planners used to spend days and weeks to manually analyze, review and validate information from disparate sources to optimize available pipeline capacity. The company wanted to improve safety and cost-efficiency of operations, create a seamless transfer of information, and provide operational recommendations to controllers for real-time optimization of pipeline performance.
Solution:
Leveraged data from existing OT systems in an Operations Data Lake, and applied Machine Learning services like Amazon SageMaker to build a forecasting model for optimizations. The solution was also able to forecast scenarios based on market conditions and provide anomaly detection and alerting for gas controllers. The company also used an intelligent document processing workflow powered with Artificial Intelligence to ingest historical paper-based data to aide with operational planning and regulatory compliance.
Impact:
- Optimization of pipeline capacity and asset utilization
- Anticipated fuel cost savings and operational efficiencies
- Processed 20M+ record images (ensure safety, maintenance, regulatory compliance)
We can now maximize capacity from our existing system to serve our customers’ needs immediately, instead of building new facilities.”
Joe Zhou
Director of Capacity Management, TC Energy
TC Energy Maximizes Operational Capacity by Innovating on AWS
TC Energy builds an intelligent document processing workflow to process over 20 million images with Amazon AI
How to get started
Phase 1: Discovery
Activities
- IT Security Review
- Data Source Identifications
- Process Flow Discovery
Outcomes
- IT Security Approval
- Finalize Data Strategy
- Infrastructure Inputs into Planning
- Define Engagement Score
Phase 2: Align
Activities
- Connectivity Identifications
- Source Prioritization
- Build RACI
- Define Models, Anomalies, User Stories
Outcomes
- Draft Architecture
- Define RACI
- Define Analytics/ML strategy
Phase 3: Launch
Activities
- Build Architectures
- Build Dashboards
- Implement and Validate Analytics/ML
- Solution Training
- End-to-End Workflow Testing
Outcomes
- Implement Solutions
- Implemented Dashboards
- Deploy Use Cases
Technology Partners
Deployment Partners
