The Swedish Defence Research Agency (FOI, Totalförsvarets forskningsinstitut) has published the report "Quantum Computing for Optimisation Problems: An Introduction with a Focus on Defence Applications", written on behalf of the Swedish Ministry of Defence. The report describes how quantum algorithms can be used to solve optimisation problems within the defence sector.

The report is written by Charlotta Bengtson, senior researcher at FOI's department for Weapons, Protection and Security, and Johan Nilsson, senior researcher at the department for Defence Technology. According to FOI's press release, the researchers examine how quantum algorithms can be applied to problems such as resource allocation and logistics.

"Optimisation problems are very common in defence calculations and in relation to resource allocation and logistics problems. It is a broad class of problems, and has at the same time been identified as a class of problems that today's quantum computers may be effective for," says Charlotta Bengtson in the press release.

According to FOI, quantum computers differ from classical computers in that they use quantum bits (qubits) instead of ordinary bits. Through the principle of superposition, a qubit can exist in both values, zero and one, simultaneously, which means a quantum computer can process multiple possibilities in parallel. Another central principle is interference, where qubits act like waves that can amplify or cancel each other out, something that is exploited to increase the efficiency of calculations.

The current generation of quantum computers is referred to as NISQ computers, Noisy Intermediate-Scale Quantum. These lack full error correction, which limits their usefulness.

"They frequently make errors, and that limits what you can do with them," says Johan Nilsson in the press release.

Charlotta Bengtson highlights that quantum computing in a defence context can provide faster solutions to time-critical problems, drawing a comparison with the aviation industry, where a near-optimal solution quickly obtained may be preferable to a perfect solution that takes a long time to compute.

The researchers note that classical computers are developing in parallel with quantum computers, and that a combination of classical computers, artificial intelligence (AI) and quantum computers will most likely be the best solution for optimisation problems going forward.

"For many problems, it is still better to use classical methods," says Charlotta Bengtson.