Posted in: Big Data Analytics, Cognitive Computing

Progress in IBM & Hartree Collaboration Reduces Physical Prototype Testing, Protects Crops from Pests and Improves Mobile Phone Coverage

The IBM and STFC Hartree team in Daresbury goes crazy for science.

More than 24 months ago, IBM and the Science and Technology Facilities Council (STFC) Hartree Centre set an ambitious goal for themselves — enable UK businesses to use modelling, simulation and Big Data Analytics on real problems to develop better products and services that will boost productivity, drive growth, increase UK competitiveness and create jobs.

To achieve this, the partners would draw on the deep expertise and know-how of their world-class researchers who are co-located at the Hartree Centre in Daresbury, UK. The team would also build upon Hartree Centre’s successful engagement with industry and its record in commercialising technological developments.

IBM’s role also includes leveraging its latest data centric cognitive computing systems and technologies, extensive amounts of background know-how and intellectual property, and the STFC/IBM teams would jointly produce reusable intellectual property assets in the collaboration that may be commercialized.

Michael Rosenfield, Vice President, Data Centric Solutions at IBM, who has been involved from the start comments, “For IBM, this collaboration has given us the opportunity to build an extremely strong and talented team in the UK, to develop useful and reusable technologies and know-how, and to leverage the engagements to drive the design of our future data centric cognitive systems.”

HRH The Duke of York officially opens the new Hartree Centre building at Sci-Tech Daresbury.

Fast Forward to Today

So where is the collaboration today? For starters, the team has hired 23 IBM scientists who with their partners at STFC Hartree, are currently engaged with more than 44 UK companies and academic and government institutions on more than a dozen proof of concept projects.

One of those projects is with EDF Energy. The partners are working together to create algorithms for performing uncertainty quantification in engineering simulations. These algorithms provide more accurate predictions about a system’s behavior (such as a centrifugal pump) and enable engineers to replace physical experiments with non-destructive virtual testing. These methods are useful to various engineering industries, from nuclear to aerospace, and allow the team to define the sensitivities of a design to uncertain conditions and predict the probability of unacceptable performance or failure. This technique can also reduce computational cost.

In a collaboration with Rothamsted Research, the scientists are using vertical-looking radars (VLR) and weather data to simulate individual insect responses to weather and insect-to-insect interactions within a swarm.

In another project the team is going back to the roots of the UK, literally. Crop damage by pests is a major threat to farming and global food security. Understanding insect behaviour and their interactions is vital in order to improve the accuracy of pest risk prediction. Knowing when and where crop damage might occur will allow the farming industry to plan preventive measures and reduce crop losses. In a collaboration with Rothamsted Research, the scientists are using vertical-looking radars (VLR) and weather data to simulate individual insect responses to weather and insect-to-insect interactions within a swarm. This approach uses high performance computing (HPC) and statistical software for quick data analysis and visualisation of results, so we can capture the dynamics that give rise to pest and crop damage risks. In addition, the use of the geospatial big data platform will have future applications in the wider area of Smart Crop Protection.

It’s become part of the standard English lexicon, “Can you hear me now?”, the dreaded words we utter when making a mobile phone call. To address this, IBM and STFC teams are modelling mobile phone connectivity using a number of open and self-collected data sources. These datasets are mined and collated using Spark on POWER8 chips and then trained. The results are provided via a web app to help travelers and businesses stay connected on all forms of transport, and while also further improving cell phone coverage by enabling smarter infrastructure decisions.

Preparing the Next Generation of UK Scientists

It’s not all work and no play at the Center. Recently, IBM, STFC’s Daresbury Laboratory (DL) and Rutherford Laboratory (RAL) Campus teamed up to host 180 girls aged 12-13, with the aim of providing a healthy and positive environment where not just women, but all the entire workforce, can prosper and fulfill their potential.

After a packed agenda with activities to introduce the girls to cutting-edge technologies with demonstrations from a number of IBM and STFC projects currently underway in the UK, they had a chance to quiz the scientists about their careers. The teams provided examples from their own personal career paths about how to get into a technology career with all the different avenues available to them, from work experience, apprenticeships, graduate schemes and professional career development.

This only represents just a small portion of the overall activities underway in the IBM/Hartree collaboration. Over the next several months the team will be demonstrating multiple proof of concept project results to multiple UK companies and are well positioned, after more than two years of very hard work, to drive these proof of concepts into production to solve real problems.

Comments

Michael Rosenfield

"this collaboration has given us the opportunity to build an extremely strong and talented team in the UK, to develop useful and reusable technologies and know-how, and to leverage the engagements to drive the design of our future data centric cognitive systems.”