Skip to main content
Home | What we do | Use cases
Tuples

Advanced algorithmic validation for aerospace logistics

A simulation and benchmarking platform to optimise loading plans for Beluga XL cargo aircraft under uncertainty.

Markets
Solutions

The challenge

Logistics planning in the aerospace sector faces extreme physical and operational constraints.

For Airbus, the European aerospace giant with over 130,000 employees worldwide, optimising the storage and loading of Beluga XL cargo aircraft is a hugely complex challenge.

Incoming and outgoing cargo flows are governed by a bidirectional, multi-queue rack system. Managing this complexity means assessing the robustness of operational plans in both deterministic and probabilistic scenarios (unexpected arrivals and departures), while ensuring the explainability of decisions, plan’s possible infeasibility and proposed alternatives.

Within the European TUPLES project, Airbus needed more than a single algorithm: it needed a scientific and technological tool to rigorously test, stress-test and validate the world’s best mathematical approaches before industrial adoption.

How we create value

Optit met this need by designing and developing Arena: an advanced competition platform. An isolated, orchestrated environment for hosting and evaluating algorithmic challenges on real industrial cases.

The platform put mathematical models from the scientific community through iterative, repeatable simulations, measuring solution quality under probabilistic uncertainty. To tackle the explainability challenge, Optit integrated Generative AI with Large Language Models (LLMs), making it possible to query the system and understand why plans were infeasible. Through this architecture, we turned a theoretical problem into an objective test bed.

Decision Science was applied to create an algorithmic governance framework, speeding up the transfer of technology from research centres to Airbus’s critical operational processes.

Impact

Adopting Arena delivered concrete benefits for organisers and partners:

  • Rigorous, repeatable validation of diverse mathematical models on complex industrial data.
  • Reduced management time for logistics challenges thanks to automatic orchestration.
  • Trustworthy AI integration via LLMs for the automated analysis of infeasibility causes.
  • Faster technology transfer into operational business systems.
Advanced algorithmic validation for aerospace logistics
See also

Efficity

Efficity
Optimising urban districts while slashing consumption and emissions.
read more
2017

A branch-and-bound algorithm for the knapsack problem with conflict graph

Bettinelli A., Cacchiani V., Malaguti E., INFORMS Journal on Computing, 29(3), pp. 457-473, 2017.
Leggi il paper

Managing uncertainty in rail networks

Alstom
Managing uncertainty in rail networks
Split-second decisions for precise journey planning.
Custom solutions
Logistics and supply chain
read more