Lead Research Organisation: STFC - Laboratories
Department Name: Scientific Computing Department


Infrastructure provides the backbone of modern societies. In our daily lives, we depend on supplies of water and energy, road, rail and other transport networks, communication networks, and waste disposal and recycling, as well as the buildings we live and work in. How these systems work has a major impact on our economic, social and physical well-being, and also on the natural environment, particularly in the light of climate change. Consequently, infrastructure investment is seen as a driver for prosperity. The UK's National Infrastructure Strategy sets out the government's plan for a renaissance in infrastructure, with hundreds of billions of pounds of public and private investment. However, the impact this will have on the economy, society and the environment, is not clearly understood. This uncertainty results from the complexity of infrastructure systems and their interaction with people and the environment.
University research supported by EPSRC has been actively developing computer models to simulate infrastructure and explore its impact, to support decision making. However, research is hampered by fragmentation of effort, and limitations of local resources to scale models to the range and detail required for the complexity of national infrastructure. Particular limitations are access to quality data and combining models from different sectors together into "system-of-systems" models.
DAFNI's vision is to allow researchers to use state-of-the-art modelling, simulation and visualisation to better inform and develop strategic thinking. DAFNI has been developed, thanks to an £8 million capital investment from EPSRC as part of the the UK Collaboratorium for Research on Infrastructure and Cities, so researchers can run models on a common computing system with a central repository of data. Using cutting edge computers, models and simulations can be scaled to greater coverage and resolution. Collaborators can use the platform as a shared workspace, and models from different sectors combined to explore their interactions using data from different sources. DAFNI also provides a legacy, a place where models and data can be used beyond the lifetime of a project.
In March 2021 DAFNI will complete its development phase funded by EPSRC, and will be ready to deliver a service to the infrastructure research community. DAFNI-ROSE under EPSRC's Resource-Only Strategic Equipment fund, will provide resources essential to make DAFNI freely available for researchers and continue its development at the cutting edge of compute and data-intensive computing. Two years of funding for DAFNI-ROSE will widen the usage and capability of DAFNI, supporting EPSRC's Engineering and related programmes. It will extend the relationships established in the development phase, so more users and projects can reap the benefits of the investment in DAFNI. DAFNI-ROSE will also look beyond academia, engaging government and industry to build collaborations.
A key group of users are research students developing skills to explore modelling of infrastructure. DAFNI-ROSE will include a programme of training to develop the next generation of Infrastructure Engineers familiar with data intensive tools for design and planning.
Digital Twins are an opportunity to transform the efficiency and reliability of national infrastructure. In a digital twin, data from real systems are fed into a running computer simulation, to predict outcomes that can then be fed back to the real system to steer its course. DAFNI-ROSE will provide a platform for a federated national infrastructure digital twin, coupling diverse simulation models that will together contribute to digital twins that are national in scope.
Infrastructure also affects a wide range of other areas, including the environment, where and how we live, our economy and our health. DAFNI-ROSE will develop collaborations across disciplines to explore the rich interaction between the built, human and natural environments.


10 25 50