Kenedict Innovation Analytics

Connected Research in Europe: An Evaluation of the EU’s Energy Research Funding

The European Union’s Seventh Framework Programme for Research and Technological Development (FP7) came to a conclusion last year and has been followed up by the new Horizon 2020 programme. Evaluation of FP7 is well underway, but so far has not focused on the actual networks of participating organisations in its various focus areas. How are participating organisations clustered, and which organisations serve as hubs in the network? Let’s take a look at FP7’s Energy Programme to find out.

Written by André Vermeij, Kenedict Innovation Analytics

The European Union has been stimulating Research & Development in its member countries through its Framework Programmes for Research and Technological Development. The seventh installment (FP7) ran from 2007 to 2013 and had a budget of just over € 50 billion, covering research funding in a total of five specific sub-programmes. The eighth programme, dubbed Horizon 2020, started this year and covers a budget of about € 80 billion until 2020.

Progress of FP7 has been monitored in annual Monitoring Reports, which contain detailed analyses of participation in the various sub-programmes from various points of view (including e.g. participation by funding scheme, organisational type, country, et cetera). The connectivity between participating organisations based on their joint project participations, however, has not been focused on so far. Mapping the networks of organisations participating in FP7 provides an answer to various interesting questions: for instance, which organisations hold key positions in their key focus areas, and can therefore be seen as knowledge hubs? In how far are clusters of organisations focusing on different technology areas connected?

To provide an initial answer to these questions, this post contains a network analysis of all organisations that have participated in FP7’s Energy Programme. The dataset used contains information on all projects funded by the EU under FP7 and is courtesy of CORDIS (the Community Research & Development Information Service). Data originate from the EU’s Open Data Portal (http://open-data.europa.eu/en/data/).

Mapping networks based on project participations
Constructing a network always starts with answering two seemingly straightforward questions: (1) what or who constitute the nodes in the network, and (2) based on what are these nodes connected? In this particular case, the nodes in the network reflect organisations that have participated in FP7’s Energy Programme. Organisations are connected when they have participated in the same project, and have thus actively worked together towards reaching the underlying project’s objectives.

An interesting example is the Twenties project (http://www.twenties-project.eu), which received € 31.8 million of funding (the largest amount in the Energy Programme) divided over 10 participating countries and a total of 25 project partners. Each of the 25 partners is thus depicted as a node in the visualisations below, connected based on their shared project participation.

The Energy Programme’s participants network
The CORDIS data shows that a total of 298 projects were funded under the Energy Programme, with total funds of approximately € 1.3 billion given out to 1959 participating organisations¹. The total amount of connections (edges) between organisations in the full network is 20,204, which shows that the average organisation in the network has collaborated with about 10 partners during the lifespan of the Programme. To reduce visual clutter and focus on the key organisations in the FP7 Energy network, the interactive visualisation below shows the largest interconnected component of the network, filtered by all organisations with a total amount of connections (degree centrality) of 20 or more. You can zoom and pan with your mouse and click any node to see its direct connections. Colours depict clusters of organisations, while node sizes reflect their amount of connections:

Based on the above, we can immediately see that the core organisations in the network are large research institutes such as Fraunhofer (Germany), CNRS (France), and the Energy Research Centre (the Netherlands), as well as universities in Denmark, the UK and France, plus France’s largest utility company EDF. When clicking these nodes, you’ll see that their connections span multiple sub-clusters. This indicates that these organisations are active in a wide variety of energy-related techologies. Of course, this is logical for leading research institutes such as Fraunhofer and CNRS based on the sheer size and scope of their activities.

Interestingly though, organisations at the periphery of the network still often have connections spanning multiple other clusters. This indicates that Energy research under FP7 has been highly multi-disciplinary, leading to an overall very dense network picture (in social network analysis parlor, network density is over 5%).

Zooming in on the largest clusters
To get a better view on the largest clusters in the above network, the visualisation below shows the red and pink clusters at the top of the network above in which Fraunhofer, the Energy Research Centre and the Technical University of Denmark play pivotal roles in terms of connections. Colours depict the six sub-clusters of organisations within these largest clusters:

Although the full network in the first visualisation is quite dense and thus a bit overwhelming at first, further zooming in on clusters such as the above helps to reduce visual clutter and more clearly identify interconnected sub-clusters. The six clusters shown above are directly relatable to individual projects: for instance, the turquoise cluster at the bottom consists of the participants of the EquiMar project (http://www.equimar.org/), focusing on tidal and wave energy.

Evaluating R&D policy through network analytics
With the Horizon 2020 Programme well on its way now, final evaluations of FP7 are being carried out as we speak. The 7th and last Monitoring Report is to be released in the fourth quarter of 2014. A network perspective such as the above may prove to be an interesting complement to the wide variety of participation indicators already utilized in the reports.

As soon as the EU releases initial data on Horizon 2020 participations as well, make sure to check back here for an analysis!

¹: Although this does not cover the total funding of € 2,3 billion (as communicated at the outset of FP7 and in various reports and Energy Programme summaries), it does provide a good proxy of project collaboration. Future versions of the CORDIS dataset may include the remaining projects as well.

2 Comments
  1. Nice Job,
    one question: I see that ENEA is indicated in two ways: Ente per le nuove tecnologie, l’energia e l’ambiente and Agenzia nazionale per le nuove tecnologie, l’energia e lo sviluppo economico sostenibile.

    As we have simply changed the name during the FP7 but the organisation is the same, my fear is that our numbers have been splitted in two thus reducing the impact on the graph.

    Could you please have a check if possible?

    Thank you in advance

    Best Regards

    Massimo Busuoli
    Head of ENEA Brux office

    • Hello Massimo,

      Thanks for your comment. I’ll take a look and get back to you when the names are merged.

      Best regards,

      André

Leave a Reply

Kenedict’s mission

To significantly improve Research & Development organisations’ decision-making by ensuring day-to-day usage of a social network perspective in the areas of Open Innovation, Competitive Intelligence, Human Resource Management and R&D Policy.

Contact Kenedict

Phone: +31 61 55 00 888

E-mail: info@kenedict.com