What Dafni can do for you

Civil engineers today use a wide range of technical, economic and environmental information in support of infrastructure decision making. That evidence is increasingly based on the outputs and insight from advanced data analytics and computer simulation models.

Data analysis and modelling provides the opportunity to explore scenarios, simulate and optimise solutions and visualise the performance of designs to inform major investment commitments.

In 2017, the UK National Infrastructure Commission report Data for the public good argued the country could spend less but deliver the same benefits to the consumer if it did more to exploit sensors, cloud computing, artificial intelligence and machine learning and put in place secure arrangements for data sharing.

National platform

The result was the Data and Analytics Facility for National Infrastructure (Dafni), a government initiative to provide a national platform for data analysis and simulation modelling for infrastructure systems. The first prototype was revealed in June 2019 at a conference at the Royal Society.

Funded with a grant of £8 million from the Engineering and Physical Sciences Research Council as part of the UK Collaboratorium for Research on Infrastructure and Cities, the facility will support academic research and provide a secure high-performance facility for practitioners in government and business.

It is being built at the Science and Technology Facilities Council’s Rutherford Appleton Laboratory in Harwell. Requirements specification and design began in 2017 and it will be fully operational by April 2022.

Infrastructure database

Dafni has five constituent parts. Its national infrastructure database will provide a secure and centrally managed platform for storing and, where appropriate, sharing national infrastructure datasets and other data needed to support infrastructure system modelling and simulation.

It will allow access to a repository of data from multiple data providers in the format required by researchers and analysts, which should make data analysis and modelling faster, cheaper and less error-prone.

Dafni’s national infrastructure modelling service will provide the user community with the ability to improve performance of existing models, reduce the complexity of creating models and facilitate the creation of complex system-of-systems models.

To encourage collaboration, avoid reinventing the wheel and drive innovation forward, Dafni will provide a central model library that houses contributions from modellers across the infrastructure community.

The national infrastructure cloud environment on Dafni will provide a high performance, flexible and scalable cloud-computing environment.

A hybrid cloud solution is being implemented that can utilise major new computer hardware investments at the Rutherford Appleton Laboratory but also ‘cloud burst’ to commercial cloud computing at times of peak demand.

Visualisation tools

Dafni’s national infrastructure visualisation suite will provide a set of tools capable of visualising large and complex datasets, including traditional visualisations (e.g. graphs and maps) and advanced technologies including virtual reality and augmented reality. This will enable improved understanding of data both in research and strategic decision making.

New visualisation hardware at universities around the country will enable analysts and other stakeholders to access Dafni’s visualisations.

Finally, a security service in Dafni will guarantee the requisite levels of security and commercial confidentiality, including authentication and authorisation services, accounts management and monitoring.

The service has been developed with advice from the National Cyber Security Centre.

This article is based on the authors’ briefing article in the latest issue (172 CE3) of the ICE Civil Engineering journal.