Our technology originates from world-class academic research carried out at École des Ponts ParisTech within the Artificial Intelligence for Air Transport Chair, led by Professor Axel Parmentier in close collaboration with Air France. The work sought to bridge two powerful but often separate disciplines — combinatorial optimization and machine learning — to tackle the complex decision-making challenges faced by large-scale industrial systems. This research led to the creation of COAML pipelines, a new methodological and software framework that integrates predictive models directly into optimization algorithms, providing more adaptive and robust solutions.

​

The approach was further strengthened through collaboration with Professor Maximilian Schiffer and his team at the Technical University of Munich (TUM), whose internationally recognized research in logistics and optimization provided key theoretical and computational advances. Together, the joint ENPC–TUM team achieved global recognition by winning the NeurIPS 2022 Machine Learning for Combinatorial Optimization Challenge, confirming both the scientific rigor and the practical efficiency of the COAML framework.

​

Between 2023 and 2025, the technology evolved through several industrial PhD projects with partners such as Air France and Renault, proving its capacity to deliver high-impact results in real-world planning and logistics contexts. Today, a dedicated and multidisciplinary team of engineers, researchers, and business experts is focused on transforming this research breakthrough into an industrial-grade software platform, consolidating tools, standardizing methods, and ensuring robustness and quality at scale.