Aircraft Loading Optimisation with QUBO
Recently, Airbus announced their winner to The “Airbus Quantum Computing Challenge” (AQCC): the team Machine Learning Reply (MLR). In the fifth challenge — “Aircraft Loading Optimisation”, they formulated the problem and its constraints into cost functions in the form of Quadratic Unconstrained Binary Optimization (QUBO) problems. These cost functions are compatible with quantum annealers, as well as other hybrid classical-quantum optimization algorithms such as Quantum Approximate Optimization Algorithm (QAOA). Then they benchmarked the model on different solvers to evaluate the performances and capabilities of current technologies.
In our case study, we will reimplement MLR’s approach in Python. Then, we can think of ways to improve their method, such as adding additional constraints and features.
We expect that most participants would finish their assignments within 8 weeks, with a 4-hour weekly commitment and one 1-hour meeting biweekly. We will host an orientation / kick-off on October 16, 2021 and the session will conclude on December 11, 2021.
Participants would work together in groups of 3 or less. In this session, we plan to have a maximum of 3 groups.
How to register?
Registration details will be shared in WeChat group. Please add FinQ’s official account to join our WeChat groups.
What we will learn
- Case study a real industry operational optimization problem;
- Understand and implement QUBO and quantum annealer;
- Participation in quantum coding development;
- Drafting professional proof-of-concept reports for quantum technology.
Prerequisites: (should take <2 hours to learn all these)
- Basic Python: Numpy. (Knowledge of Qiskit recommended.)
- Basic linear algebra: Matrix multiplication, trace, partial trace etc.
- Basic graph theory: Nodes, edges etc.
- October 16, 2021: Orientation / kick-off
- Week 1-2: Reading materials, Q&A
- Week 3-4: Implementation & iteration
- Week 5-6: Brainstorming & adding features
- Week 7-8: Review & report composing
- December 11, 2021: Final presentation