Technology

Optimisation

Imagine a sudoku puzzle. Solving it requires systematically working round all the options until each square and eventually the whole grid is resolved. A classical computer may achieve this faster than the human brain but will tackle the problem in fundamentally the same way. By contrast, the quantum mechanics properties of qubits (quantum processing bits) mean that all possible values for each square can be represented simultaneously, and all potential combinations across the whole grid can be simultaneously understood. The speed to resolve the puzzle will be vastly increased by quantum processing compared to classical processing.

Taking a more commercial challenge, imagine a delivery schedule with many different route options to deliver to a large number of addresses, plus variables around traffic flows, fuel efficiency and weather conditions. Or consider a fund that could be invested in many different asset classes, with different risk levels and growth potential to consider not just for each class but for each investable asset within the class. The simple question of the optimal route for the driver to take or the best combination of investments requires a powerful classical computer to find an answer – but can be resolved much faster with quantum processing. Other optimisation problems include production or process planning in manufacturing, traffic-flow management and supply-chain logistics across a wide range of industries or staff scheduling.

Contracting considerations

The vast majority of quantum computing for commercial applications is accessed as a cloud service, rather than purchased for on-premise use. The early stage of understanding means that partnerships between the quantum processing service provider and end-users, including consultancy and support from the service provider, are typically critical to success.

Given that these partnerships are very often exploratory in nature, the terms and conditions for quantum processing services may be more open to negotiation than classical cloud services and require a different approach to operations and risk allocation than mature and standardised services. In particular, it is worth taking care to ensure the following are addressed:

• The scope, objective and limits of the joint project or the service that is provided should be clearly defined or, if not, mechanisms for refining the scope should be agreed.
• Given that quantum processing capabilities are likely to be an important component of the arrangement, relevant metrics for these should be considered.
• The nature of the relationship and the specific project should be reflected in an appropriate pricing mechanism that aligns the parties' interests.
• Risks, responsibilities and liability should be clearly mapped out and understood, particularly in relation to errors generated by the quantum system, paying attention to the potential for errors that may be hard to foresee due to the complexity of quantum computing technology.