Quantum computing: how can it serve your organisation?

Practicable algorithms for quantum optimisation

The quantum computer is increasingly in the news and is portrayed as having great promise. But what can you actually do with it? What problems can you solve with quantum computing applications? And how can your organisation make the best use of them in the future?

Computing power provides opportunities

Quantum computing provides opportunities for solving all sorts of problems that current technology cannot deal with for lack of computing power. Yet quantum computers will not completely replace today’s technology. Current and quantum technologies are expected to go hand-in-hand. This will open the door to all kinds of applications, especially in the following fields:

  • optimisation and simulation
  • AI (Artificial Intelligence)
  • materials science.

Companies in sectors such as automotive, logistics, finance, pharma, and chemicals are already encountering problems that are practically unsolvable with current computers.

A good solution to optimisation problems requires a lot of calculation time. This is where quantum computers could help significantly with tasks such as:

  • improving traffic flows in real time, so that there are fewer traffic jams and delays
  • finding shipping routes that are cheaper and produce lower carbon emissions
  • optimising production processes
  • optimising portfolios for financial institutions.

Quantum computing applications in the Netherlands

A quantum computer can be used as a large parallel computer with enormous computing power. It can evaluate many potential solutions simultaneously and thus find a good solution more quickly. At TNO, for example, we have done research on the use of quantum computing for inland waterway routing, portfolio optimisation, and the placement of access points for wireless networks.

The quantum computer can also behave like a neural network. This is a widely used technique in machine learning, which is a part of AI. In some cases, applications can be trained more quickly, display more complex relationships, or accommodate more patterns. We’ve studied various quantum computing applications:

  • reinforcement learning
  • support vector machine implementations
  • capacity increase of Hopfield networks
  • quantum classification methods for mobile phone positioning.

The huge computing power of the quantum computer makes it possible to simulate quantum processes and analyse complex molecular structures for materials science. This provides many different opportunities in various fields:

  • efficient fertiliser development
  • better batteries and other materials
  • medication tailored specifically to your body.

Opportunities for Quantum and AI

A quantum computer makes smart use of quantum mechanical effects. This can result in exponentially faster algorithms. And it can potentially find solutions to specific complex problems, which will provide time savings. This also applies to using quantum technology to complement the capabilities of AI. It’s expected that quantum algorithms will be a widely used tool in AI and optimisation in 10 to 20 years’ time. Even in the shorter term, these algorithms can help with specific problems.

Simulating algorithms to solve use cases

At TNO, we’re working on quantum algorithms to demonstrate the added value of quantum computers. Current quantum computers are still too small to provide the desired computing power. Therefore, many algorithms are simulated. We do this with the aid of Quantum Inspire, the quantum computing platform developed by QuTech , with a quantum simulator and two types of quantum computers, among other things. This platform has been designed to support the first steps in the fascinating world of quantum computing. And to explore the possibilities for using the power of quantum computers in the future.

We’re working on various use cases, from anomaly detection to route optimisation. And we’re trying to gain insight into the infrastructure needed for the use of quantum computers (in the medium and long term) in the most promising application areas.

Sample case on routing strategies

In one case, we looked at optimising routing strategies. The technique we used for this is called reinforcement learning, a form of machine learning. Existing literature only takes account of the starting point of one individual, the desired destination, and possible obstacles along the way. We’ve extended this to a setting with several departing individuals. It uses what is called a quantum annealer from D-Wave. This special quantum system specialises in solving optimisation problems. We’ve used it to show that it may lead to time savings by finding a good route faster.

Currently, a limited problem size can be processed. The quantum hardware still needs to be developed further in order to solve practical problems. But once we’ve reached that point, it will have an impact on many real-life scenarios, such as:

  • stationing of military troops
  • efficient delivery of post and parcels
  • applications in warehouses and traffic.

Want to know more about quantum computing?

Would you like to know more about our activities in the field of quantum algorithms? Or do you want to work with us on a use case? Please contact Frank Phillipson.