NASA
National Aeronautics and Space Administration
NASA Sustainable Aviation Operations
Reducing the impact of aviation on our planet
By
Dr. Pankaj Dhussa
Driving Behavioral Change for Information Management through Data-Driven Gree...
NASA Sustainable Aviation Operations Reducing the impact of aviation on our planet
1. Sustainable Aviation Operations
Reducing the impact of aviation on our planet
Challenge
• Commercial airplanes and large business jets
contribute approximately 10% of the United States
transportation emissions
• There is a need to reduce the impact of aviation on
our planet to achieve the United States Climate
Action Plan objective of net-zero emissions by 2050
• The lack of access to information and the lack of a
digital workflow for rerouting constrains the ability
to identify and implement trajectory changes that
improve sustainability
Improved access to digital information enables
aviation services that improve sustainability
A digital trajectory negotiation workflow enables
implementation of sustainable trajectories
Expected Impacts
• NASA’s ATM-X project is improving the
sustainability of aviation operations to provide
fleet-wide benefits to existing aircraft
• Improvement to sustainability will reduce the
impact of aviation on our planet by:
— Reducing emissions
— Reducing fuel use
— Reducing persistent contrails
Partners
• The Federal Aviation Administration (FAA)
• Aviation service providers
• Airline partners
Solution
• Develop a reference digital information platform to
identify requirements for sharing digital information
needed to train and implement predictive machine
learning services
• Develop and demonstrate machine learning services to
improve predictions of the airspace and identify
sustainable reroute opportunities
• Demonstrate a digital rerouting workflow that enables
operators to seamlessly implement sustainable routes
identified aviation services
Results
• Using machine learning can help scale sustainable aviation
services across the National Airspace System by enabling them
to learn about local airspace adaptations
• Operational evaluations showed real world sustainability
benefits
Machine learning services improve predictions of
the National Airspace System
University challenges empower the next generation
to solve challenging real-world problems
Next Steps
• Scale operational demonstrations of predictive services to
Houston to validate scalability to different airports
• Flight demonstrations of a digital rerouting workflow that
incorporates FAA, airline dispatch, and connected flight deck
systems
• Extend predictive services to improve the resilience of fleet-
wide irregular operations
Challenge Goal:
Phase I: Build a model to predict pushback time at US Airports
Phase II: Build a model based on Federated Learning framework
Five Winning Teams out of 458 Submissions
SFNP Ops 2
FY23-27
Integrated
Airborne
Rerouting
Fleet Wide
Irregular
Operations
Management
SFNP Ops 3
FY24-28
SFNP Ops 4
FY25-30
Capstone
Demonstration
Collaborating with the FAA and industry to demonstrate
reduced emissions and fuel use for aviation operations
Digital Pre & Post Departure
Rerouting Workflow and and
Contrail Avoidance
Aviation Services to Optimize
Pre Departure Routing
SFNP Ops 1
FY22-25
Pre Departure
Rerouting
`
4D trajectory optimization through
an end-to-end digital workflow
Fleet Optimization for
Irregular Operations
Machine Learning Based Services
Automated Scalable Surface Operation Prediction Model
Absorbs information from
multiple existing data sources
Flexible to new tasks
Learns detailed knowledge
Semi-automated
maintenance
SME data
Surface data
Similar to:
3D printer
Updating…
5
Machine learning is being used to achieve real world benefits and to inform operational decisions
Services currently deployed include:
• Airport configuration prediction
• Taxi time prediction
• Runway prediction
5
Real-World Sustainability Benefits (2022-2023)
Fuel Savings
Actual: 68 flights
Over 55,000 lbs
Emissions Savings
Actual: 68 flights
Over 169K lbs. CO2 or
over 1200 urban trees
Learn More