NOADS#
Numerical Optimization of Aviation Decarbonization Scenarios with GEMSEO-JAX
NOADS is a Python framework for finding optimal decarbonization pathways for aviation. It combines multidisciplinary optimization (GEMSEO) with automatic differentiation (JAX) to explore trade-offs between fleet modernization, alternative fuels, and demand-side policies, all within the carbon budgets defined by the IPCC’s 6th Assessment Report.
NOADS serves as a proof of concept for advanced numerical methods (automatic differentiation, JIT compilation, vectorized multi-scenario optimization) applied to aviation decarbonization scenario analysis. The models build on the AeroMAPS platform developed at ISAE-SUPAERO, and the numerical methods developed here are intended to be re-incorporated into AeroMAPS. For the full scientific context, see the companion paper, available in this documentation as an online extended paper:
Costa-Alves I., Gourdain N., Gallard F., Gazaix A., Kambiri Y.-A., Druot T. (2026). Numerical optimization of aviation decarbonization scenarios: balancing traffic and emissions with maturing energy carriers and aircraft technology. Applied Energy, 412, 127631. https://doi.org/10.1016/j.apenergy.2026.127631
BibTeX
@article{costaalves2026,
title = {Numerical optimization of aviation decarbonization scenarios: balancing traffic and emissions with maturing energy carriers and aircraft technology},
journal = {Applied Energy},
volume = {412},
pages = {127631},
year = {2026},
issn = {0306-2619},
doi = {https://doi.org/10.1016/j.apenergy.2026.127631},
url = {https://www.sciencedirect.com/science/article/pii/S0306261926002837},
author = {Ian Costa-Alves and Nicolas Gourdain and François Gallard and Anne Gazaix and Yri Amandine Kambiri and Thierry Druot},
keywords = {Multidisciplinary optimization, Low-carbon fuels, Aircraft design, Integrated assessment models, Shared socioeconomic pathways},
}
About this documentation
This documentation site, and the reorganization of the repository around it, were prepared with the assistance of Anthropic’s Claude (Opus 4.8): structuring the pages, wiring the example galleries, drafting the getting-started, extending, and API material, and migrating the build from MkDocs to Sphinx. The extended paper reproduces the authors’ own manuscript and supplementary information verbatim, and the scientific models and results are entirely the authors’ work.
Install NOADS, learn the core concepts, and run your first scenario optimization in 10 minutes.
The companion paper in web form: models, results, supplementary material, and the runnable examples that produce every figure.
Partition the fleet differently, add aircraft, pathways, or energy resources, and quantify uncertainty.
The complete API documentation generated from the source docstrings.
Key capabilities#
Fleet and aircraft design: model current and future aircraft across market segments (commuter to long-range); new aircraft are sized with the Generic Airplane Model (GAM) under evolving technology assumptions.
Energy production system: trace fuels from primary resources (biomass, electricity, crude oil) through production pathways (HEFA, Fischer–Tropsch, electrolysis) to final energy carriers, with full CO₂ and cost accounting.
Temporal scenarios: optimize policy trajectories from 2025 to 2075 with time-dependent controls, ODE dynamics for technology diffusion, and cumulative constraint budgets.
Climate scenario integration: drive scenarios with IPCC AR6 data: SSP pathways for GDP and population, carbon budgets for 1.5°C to 2.5°C targets, and biomass/electricity availability bounds.
Fast gradient-based optimization: JAX automatic differentiation provides exact gradients, enabling speedups of two orders of magnitude over finite-difference approaches.
Robust and multi-objective policies: find policies that are robust across multiple climate futures, or explore Pareto trade-offs between emissions and cost.
Quick example#
from noads.application.examples import single_policy_scenario_optimization
results = single_policy_scenario_optimization(
global_scenario_name="SSP2-26", # SSP2 pathway, 2°C target
technology_index=1, # mid-tech assumptions
carbon_budget_percent=3, # aviation's share of global CO₂ budget
plot_optimum=True,
)
Credits#
The Generic Airplane Model (GAM) was re-written in JAX together with Yri Amandine Kambiri. See Kambiri et al., Energy consumption of Aircraft with new propulsion systems and storage media, SciTech Forum, Orlando, January 2024.
Open-access data sources used for calibration:
World Bank: Population, GDP, air transport departures (World Development Indicators).
ICAO: Revenue Passenger Kilometers (via Airlines for America).
IPCC AR6: Primary energy, electricity, emissions, GDP, and population projections (AR6 Scenario Database, Byers et al., 2022).
License#
Source code: GNU LGPL v3 (GPL v3 for AeroMAPS-derived equations)
Examples: BSD 0-Clause
Documentation: CC BY-SA 4.0