Engineering and Physical Sciences Research Council Research outputs 2016

Transcription

1 Engineering and Physical Sciences Research Council Research outputs 2016 An overview of outcomes and impacts of EPSRC-supported research

2 Contents Introduction 1 Executive summary 2 Research grants section 4 Key findings 4 Narrative impact 5 Publications 7 Collaborations and partnerships 10 Further funding 15 Facilities 18 Next destination and skills 20 Influence on policy and practice 22 Physical outputs summary 24 Intellectual property and licensing 25 Spin-outs 27 Engagement activities 29 Awards and recognition 31 Secondments 33 Students section 34 Key findings and narrative impact 34 Publications 36 Collaborations and partnerships 37 Facilities 39 Next destination and skills 41 Physical outputs summary 42 Intellectual property and licensing 43 Spin-outs 44 Engagement activities 45 Secondments 47 Case studies 48 Annex: Compliance 52

3 Acknowledgement EPSRC is grateful to all the researchers and students who provided an up-to-date record of the research outcomes used in this report. Their support is very much appreciated and will help us to understand and to demonstrate the benefits of investing in engineering and physical sciences research and training.

4 Introduction This report summarises the outputs, outcomes and impacts so far attributed to current/recent EPSRC funding and submitted to EPSRC using the researchfish system 1,2. In common with most other research councils, 2016 was the second time EPSRC used Researchfish to collect this information from research grant principal investigators 3 ; it was also the first time we used it to collect the same information from research students supported by EPSRC block training grants. EPSRC uses the data gathered through Researchfish in a number of ways, examples include: in regular reports to BEIS; to inform research council strategy and policy; as an input to evaluations; to contribute to larger studies drawing on outcomes data from multiple funders. examples in this report include the use of standard industry codes to categorise the industry sectors in which EPSRC research has given rise to spin-out companies, and the further analysis of intellectual property. It is difficult to predict research outcomes in advance, and researchers are not required to report a minimum number of any particular type of outcome. While this reduces the reporting burden it also inevitably leads to under-reporting across many types of outcome. The Researchfish data reported here therefore needs to be interpreted with care it does not represent the sum total of outcomes and impacts arising from EPSRC-funded research. Rather, the aggregates reported should be seen as minima which reflect the distribution and relative volumes of different outcome types. The report is in two main sections: the first covers the outcomes arising from research grants 1 and fellowships, the second covers the outcomes reported by research students 2. Often the data is used in combination with data from other sources to paint a richer picture 1 The report covers outcomes of grants/fellowships which were either current at the start of 2016 or which had ended after 1 November Most data was provided during the 2016 submission period (01/02/16 to 17/03/16), but some was provided during the 2014 submission period (01/10/14 to 30/11/14) and not updated in The report also covers research outcomes reported by research students as arising from the research undertaken during their studies. All such data was collected during the 2016 submission period from students who had either completed their studies since 2013 or, if still studying, had commenced their studies earlier than 30 September The majority of the data was provided by students still engaged in study; it is likely this reflects a loss of contact with many students after they leave university. 3 MRC and STFC used Researchfish for several years before it was adopted and used by all the research councils for the first time in

5 Executive summary EPSRC, in common with the other UK Research Councils, uses the Researchfish online system for gathering information on research outputs and outcomes. The data collection period held in early 2016 was the second time we asked researchers to use the system to provide an update on their research outcomes, and was the first time we asked research students supported by our training grants to do so. This report provides an overview of the data gathered about outcomes arising from EPSRC grants with end dates after 1 November Outcomes information was provided on 98% (by value) of the EPSRC grants on which reporting was required. Framework and strategic university compliance rates varied between 96% and 100%, with the majority achieving 100%. The outcomes of EPSRC-supported research are relevant to a wide range of sectors: energy, digital/communication/information technologies and healthcare are most often cited as sectors of potential interest and where impacts have already been delivered, but many other sectors are also cited significantly frequently. Almost 100,000 publications have been reported, making them the most commonly reported output: 87% of grants report at least one, and they account for 50% of all reported outputs. Journal articles are the commonest type of reported publication (71%). Significant numbers of other outputs have also been reported, including: software and technical products (1,408); research tools and methods (652); research databases and models (685); artistic/creative outputs (487); products and interventions (90). Almost 50% of EPSRC grants (by value) report at least one collaboration or partnership with research users, of which some 85% are with private/voluntary/civic sector organisations. Half of all collaborations reported through Researchfish are described as new; and international collaborators are most often based in the USA. Approximately 6.5bn of further funding has been reported by roughly a third of awards, representing approximately 2.7 times the value of the grants reporting the funding. One fifth comes from overseas, and the main sources are academia ( 3.3bn) and the public sector ( 2.4bn). The majority of reported next destinations of researchers relate to post-doctoral researchers and research students, and show that: approximately two thirds moved on to another post in academia/other research council; approximately one fifth took up a role in the private sector. Over 900 policy influences have been reported, demonstrating the significant relevance of EPSRC funded research to a wide range of sectors from energy and environment through communications, healthcare, and transport, to security and the creative economy. Influence through participation in an advisory committee is most commonly reported, followed by training of practitioners/researchers, and membership of a guidance committee. Data from Researchfish and in legacy systems shows that EPSRC-funded research has resulted in more than 1,000 patent applications in the past decade; approximately one half have been granted and one third are still under examination. The Researchfish data shows that 80% of reported intellectual property (IP) is protected by the filing of patent applications. The data submitted to Researchfish includes reports of 239 spin-out companies, established to exploit the outcomes of EPSRC-supported research. Drawing on additional data from other sources, EPSRC is aware of at least 548 spinout companies created since 2004, of which 80% are still active. The data shows that grants considered relevant to electronics, healthcare and manufacturing are most likely to report a spin-out company, and that EPSRC research stimulates innovation and activity across a wide range of industry sectors. 2

6 More than 15,800 instances of engagement have been reported. Almost irrespective of the audience, the commonest method is through giving a talk or presentation (47% of the total), followed by participation in workshops or similar activities (26% of the total). Most of the engagement activity (36%) has been directed towards professional practitioners (engineering professionals, for example) and just over 20% has been delivered with the public as the primary audience. Overall, these data demonstrate the wide range of outcomes and impacts arising from EPSRCsupported research 4. As we continue to build on this with further data we will be able to demonstrate more fully the value of investing in engineering and physical sciences for the UK. EPSRC-funded research leads to career prestige in at least 20% of EPSRC awards. More than half the reported instances cite personal invitations to serve as a conference key note speaker. The next commonest form of recognition is the award of a research prize (17%). 4 Caution is advised when drawing conclusions from the data in this report. Although principal investigators are required to submit up to date reports of their research outcomes, and may lose their eligibility for research council funding if they fail to do so, they are not required to submit a minimum number of any particular type of outcome. This leads to an inevitable degree of under-reporting, meaning that the aggregated data in this report is therefore generally indicative of the lower boundary of the actual volumes. 3

7 Research grants section Key findings Energy, digital/communications and healthcare are the most frequently relevant sectors. The key findings section offers an opportunity for researchers and students to report on the main highlevel results from the award, particularly in relation to their objectives. They are asked to describe for a non-specialist audience, what the most significant achievements were from the award and how the findings might be taken forward and by whom. The main purpose is to enable this information to be shared with other researchers and research users, for example, via the research councils Gateway to Research, but analysis also provides insight into the overall relevance of the research. In order to understand the influence of the research, researchers and students are asked to suggest one or more sectors to which their findings might potentially be of interest. Of the 7,767 grants on which key findings were reported in 2016, the most commonly identified sectors of relevance were energy (31.6% of the awards reporting key findings), digital/communication/information technologies (29.7%) and healthcare (25.1%) (see Figure 1). Figure 1: Key findings according to sector of potential interest Energy Digital/communication/information technologies Healthcare Environment Aerospace, defence and marine Chemicals Electronics Manufacturing, including industrial biotechnology Pharmaceuticals and medical biotechnology Education Transport Other Culture, heritage, museums and collections Agriculture, food and drink Construction Security and diplomacy Creative economy Communities and social services/policy Government, democracy and justice Financial services and management consultancy Leisure activities including sports, recreation and tourism Retail ,948 1,792 1,718 1,659 1,617 1,525 1,393 1,161 1,080 1, ,308 2,458 0% 5% 10% 15% 20% 25% 30% 35% Percentage of total awards reporting key findings (number of awards to right of bar) 4

8 Narrative impact Over half of research grants report having economic impact, most frequently in energy, digital/ communication and healthcare. Significant levels of societal, cultural and policy impacts are also reported. In the narrative impact section, researchers summarise the evolving overall economic and societal impact of the work supported by the grant. It is accepted that impact can occur over a wide and varying timeframe, and the section can therefore be updated as impact emerges and develops. It is expected that it will begin to be possible to identify impact a year after an award has ended, and that updates will continue to be provided for at least five years after the end of the award. Researchers are asked to report how the findings are being, or have been, used beyond academia in the public, private, third/voluntary sectors and elsewhere. The information will be shared with other researchers and research users via the research councils Gateway to Research. Of the 5,490 grants on which narrative impact was reported in 2016, the most commonly identified impact type was economic, reported by 53.9% of grants. The most frequently reported type of impact in 2014 was also economic (41.5%). Significant levels of other types of impact were also reported (see Figure 2). (N.B. Researchers were able to select multiple types of impact). Figure 2: Types of impact reported in narrative impact Economic Societal Cultural Policy & public services ,000 1,500 2,000 2,500 3,000 3,500 Number of awards The sectors in which impacts most commonly occurred in 2016 are: energy, digital/communications, and healthcare for awards a very similar distribution to the areas of potential interest identified by researchers in the key findings section (see Figure 3). 5

9 Figure 3: Sectors in which impacts have been achieved Energy Digital/communication/information technologies Healthcare Education Environment Aerospace, defence and marine Manufacturing, including industrial biotechnology Chemicals Electronics Pharmaceuticals and medical biotechnology Transport Other Culture, heritage, museums and collections Construction Agriculture, food and drink Creative economy Security and diplomacy Government, democracy and justice Communities and social services/policy Financial services, and management consultancy Leisure activities, including sports, recreation and tourism Retail , ,227 1,364 0% 5% 10% 15% 20% 25% 30% Percentage of total awards reporting narrative impact (number of awards to right of bar) 6

10 Publications Publications are the most commonly reported output: 87% of grants report at least one, and they represent 50% of all reported outputs. Journal articles are the commonest type of reported publication (71%). In this section researchers are asked to record brief details of their publications and to attribute them to the grant(s) which supported the work being reported. Approximately 4,000 publications were harvested from external sources and automatically recorded by Researchfish - for example where EPSRC funding is already correctly acknowledged and recorded in an external database this information can be harvested by Researchfish. A further 1,000 were provided in bulk datasets by research organisations on behalf of their researchers as part of the Researchfish-RCUK interoperability project. Publications are the most frequently reported research output. They account for over 50% of all outputs attributed to EPSRC grants, and approaching 90% of grants report at least one publication. Table 1: Publications summary Publications summary Percentage of group reporting at least one publication Number of awards in group Number reporting at least one publication Total number of publications reported by the group Total number of unique publications reported by the group Awards 87% 9,181 8, ,890 99,716 The journal article is the commonest type of publication, comprising 71% of all publications attributed to awards. Most journal articles have unique identifiers and duplicate records created by different researchers can be easily distinguished. The same publication may legitimately be reported as an outcome by several different grants. The unique count uses the Digital Object Identifer (DOI) where possible to identify each publication so that it is only counted once in the total. 7

11 Figure 4: Distribution of publication types 2% 1% 10% 16% 71% Journal article Conference proceedings abstract Other Book/chapter/book edit Technical report/standard Thesis/other student publication Consultancy/policy report ( Other includes the following categories selectable by researchers when entering details of a publication manually. Lecture-speech, manual/guide, monograph, newsletter-article, online-resource, other) The extent to which academic journal articles are cited is widely accepted as a proxy indicator of the quality of the work being reported. Citations build over time: publications in a particular field and of a particular age can be expected, on average, to have been cited a certain number of times, and field - and publication year - weighted percentiles can be calculated. Analysis of journal articles attributed to EPSRC-funded research and published between shows that they are generally two to three times as likely to be in the top percentiles as other papers published in the same field and year, as shown in figure 5. For example, the green column shows that among articles published in 2012, those arising from EPSRC funding are two and a half times more likely to be in the top 5% of articles by count of citations gathered since publication. 8

12 Figure 5: Likelihood of EPSRC articles appearing in top citation percentiles 4 Normalised occurrence of journal articles arising from EPSRC grants in field weighted top 1%, 5%, 10% and 25% percentiles (Citations since publication as at July 2016 Source: Elsevier SciVal ) 25% normalised 10% normalised 5% normalised 1% normalised Likelihood of being in the indicated percentile (1 = world average) Publication year Note: Because citations build over time, the above chart indicates not only that publications arising from EPSRC-funded research are rapidly cited significantly more often than world average, but that they tend to grow in relative importance as time passes. Thus while articles published in 2015 are already cited so often that they are approximately 2.3 times more likely to be in the top 5% than world average, articles published five years earlier have accrued citations faster than others and are approximately 2.7 times more likely to be in the top 5%. 9

13 Collaborations and partnerships Almost 50% of EPSRC grants (by value) report collaboration with research users. Some 85% of these are private/voluntary/civic organisations. Half of all collaborations reported through Researchfish are described as new ; international collaborators are most often based in the USA. For this section researchers are asked to report any new or changes to existing collaborations. Also to be recorded here are details of any bi-lateral or multi-lateral partnerships, participation in networks, consortia or other initiatives and collaborations with other departments or researchers within their institution. It was requested that collaborations at an early stage of discussion, where there has as yet been no tangible output or where collaboration details are restricted by contractual confidentiality, should not be included. Table 2: Collaborations summary 5 Collaboration summary Number of grants in group Value of grants in group Percentage of group by value of grants reporting at least one collaboration via Researchfish Percentage of group by value of grants reporting at least one collaboration with any organisation at proposal stage and/or via Researchfish Percentage of group by value of grants reporting at least one collaboration with a research-using organisation at proposal stage and/or via Researchfish Number reporting at least one collaboration via Researchfish Number reporting at least one collaboration at proposal stage and/or via Researchfish Total number of collaborations reported by the group Total number of unique collaborations reported by the group All 9, bn 35% 57% 48% 2,675 4,426 10,876 9,765 It is worth noting that details of intended collaborations are submitted with grant proposals; often these details have been updated in Researchfish, and information about new collaborations has also been provided. However, some researchers have not updated the information submitted with their grant application. To fully appreciate the extent of collaborative activity it is therefore necessary to consider data from Researchfish and data from grant proposals 6. 5 While most sections in this report refer to 9,181 grants, this table excludes 79 grants which included funding for capital equipment that would have significantly distorted the subsequent analysis. 6 Of the 9,102 awards, 33% reported one or more collaborations at the proposal stage, but over half of these did not report further details in Researchfish. Meanwhile, 1,400 other grants, which did not include details of intended collaboration at the proposal stage, did report details of a collaboration via Researchfish. 10

14 The data so far submitted to Researchfish on the value of collaborators contributions is insufficient to allow direct comparison with grant values; however, the available data confirm that: 57% (by value) of EPSRC grants submitting outcomes to Researchfish have attracted collaboration and/or co-funding support from all sources; 48% (again by value) have attracted such support from research-using organisations 7 ; 11% of collaborating research-using organisations are public sector, typically within the NHS or national/local government. Details of new collaborations, defined as when the year the collaboration starts is later than the start year of the grant, have been reported in 14% of awards. Figure 6 illustrates the geographical distribution of collaborations reported for EPSRC grants in Researchfish. The majority of the collaborators were from the United Kingdom (46.5%) followed by other (29%) - this includes countries other than USA, Canada, China, Japan, Australia, UK, EU - and then the EU (12%). The size of the pie chart shows the relative percentage of collaborations (by number). The divisions within each pie chart represent the extent to which researchers and students are engaging with collaborators from different sectors at that location. 7 Research-using organisations include public, private, charitable and civic organisations, and exclude universities and research council-funded institutes. 11

15 Figure 6: Number of collaborations by location and sector (awards) Note: Hospitals includes NHS and hospitals overseas; Multiple includes joint collaborations; Other includes collaborations not mapped to sector. 12

16 A list of the top 15 private sector collaborators reported in Researchfish can be seen in Table 3. Table 3: Top 15 private sector collaborators (awards) by number of collaborations Collaborator Rolls Royce Group Plc BAE Systems AstraZeneca Airbus Group Arup Group Tata Steel Europe Johnson Matthey GlaxoSmithKline (GSK) Network Rail Ltd ARM Holdings Microsoft Research EDF Energy Renishaw Plc Depuy International QinetiQ The UK private sector collaborations were mapped to industrial sector by Standard Industrial Classification (SIC) code and this showed that the highest number (33%) were in manufacturing, followed by professional, scientific and technical (27%) and information and communication (11%) (see Table 4). 13

17 Table 4: Industrial sector of private sector collaborations Industry Count % Manufacturing Professional, scientific and technical Information and communication Wholesale and retail trade, repair of motor vehicles and motorcycles Administrative and support services Construction Electricity, gas, steam and air conditioning supply Transportation and storage Human health and social work Other services Public administration and defence Water supply, sewerage and waste management Financial and insurance activities Arts, entertainment and recreation Education Mining and quarrying Accommodation and food services Real estate activities Agriculture, forestry and fishing Total , The 976 collaborations from the manufacturing sector were broken down further into subtypes. It was found that 30% were from the manufacture of computer, electronic and light electrical goods and 22% were from the area of machinery, vehicle and transport manufacturing. 14

18 Further funding Approximately 6.5bn of further funding has been reported by roughly a third of awards, representing 2.7 times the value of the grants reporting the funding. One fifth comes from overseas, and the main sources are academia ( 3.3bn) and the public sector ( 2.4bn). For this section researchers are asked to record any further funding received as a result of the research supported by the award and to include any funding awarded by the research councils or other funding providers. It can also include scholarships, studentships, fellowships or travel awards. Table 5: Further funding summary by instances Further funding summary by instances Percentage of group reporting at least one instance of further funding Number of awards in group Number reporting at least one instance of further funding Total number of instances of further funding reported by the group Total number of unique instances of further funding reported by the group Awards 33% 9,181 3,070 10,872 9,928 Table 5 shows that researchers reported instances of further funding in 33% of the 9,181 EPSRC awards. The total value of further funding for the awards reporting further funding outcomes is 6,473m. The total grant value of these 3,070 awards is 2,363m - achieving a gearing of around 2.7. The sectors providing the largest value of further funding from researchers were the academic sector ( 3,291m, 50.8% of the total value of further funding reported) and the public sector ( 2,411m, 37.3%) (see Table 6). Table 6: Further funding by source sector and reported value Source sector Reported value ( m) % Academic/university Public Charity/non-profit/hospitals Private Other Learned Society Multiple Total 3,291 2, ,

19 The majority (52.9%) of the public sector further funding came from overseas (Europe/EU). The research/funding councils, TSB/Innovate UK contributed 15.4% and the UK Government 11.8% of the public sector further funding total (see Table 7). Table 7: Further funding public sector location by value Further funding public sector location group Further funding value ( m) % Overseas (Europe/EU) Research/funding councils, TSB/Innovate UK UK Government UK NHS/health UK public sector body Overseas (USA) Overseas (rest of World) UK regional/local Total 1, , The map in Figure 7 illustrates the location where the further funding originated, based on country of origin of the organisation providing funds. For the UK and the EU, the size of pie charts indicates the relative percentage of further funding value: 75% of the further funding total came from the UK, 19.5% from the EU. For the other locations on the map (USA, China, Japan, Australia, other), the size of the plain coloured bubbles indicates the relative percentage of further funding value. The divisions within each pie chart represent the proportion of further funding value from different sectors at that location. 16

20 Figure 7: Value of further funding by location and sector 17

21 Facilities EPSRC grant holders reported use of over 170 different facilities to support research funded by 530 research grants. Over half (56%) of these facilities are in the UK, with 25% being elsewhere in Europe, 12% in the USA and the remainder (7%) in other countries. This section captures instances when EPSRC funded research was made possible through the use of specialised research facilities such as beamlines to study material properties or high performance computers to run complex systems models. It can be difficult to determine when specialised equipment is a facility and when it is not. The data collected not only show how research is enabled by accepted facilities such as Diamond and ARCHER, but also the important role played by the sharing of highvalue research equipment within and between research organisations. Table 8: Use of facilities summary by instances Percentage of group reporting at least one instance of facility usage Number of awards in group Number reporting at least one instance of facility usage Total number of instances of facilities usage reported by the group Total number of unique instances of facilities usage reported by the group 6% 9, , EPSRC grant holders reported use of over 170 different facilities to support research funded by 530 research grants. A little over half (56%) of these facilities are in the UK, with 25% being elsewhere in Europe, 12% in the USA and the remainder (7%) in other countries. Table 9: Location and number of facilities used, and volume of grants reporting facility use Location UK Rest of EU USA Rest of World Total Count of facilities Unique reports of use by research grants

22 The figures in Table 10 are the numbers of research grants which reported use of the named facilities. The figures are indicative rather than absolute because summarising the data is difficult, due to lack of adoption of standardised names and the frequent use of a site location name as a proxy for the actual facility used. Of the 27 reported instances in which other HPC facilities were used, 21 made use of UK facilities mainly based in universities. Table 10: Top 10 facilities most frequently used to support EPSRC-funded research Facility DIAMOND Synchrotron STFC ISIS Pulsed Neutron and Muon Source ARCHER/HECToR Institut Laue-Langevin Neutron Scattering Facility STFC Central Laser Facility European Synchrotron Radiation Facility (ESRF) Other HPC Facilities Swiss Light Source (SLS)/Paul Scherrer Institute (PSI) EPSRC National Service for Mass Spectrometry UKCCSRC Carbon Capture PACT facilities Research grants

23 Next destination and skills Two thirds of researchers (mostly post-doctoral researchers) employed on EPSRC grants remain as active researchers in academia and nearly one-fifth (19%) take up a role in the private sector. The next destination section is used by researchers to record details about staff who have left their team and where their salaries were wholly or partly funded by their grants. The information includes the role of the member of staff before they left and the role, location and sector of their next destination. The percentage of awards which provided details of at least one next destination output was 35% for awards. (see Table 11). Table 11: Next destinations summary by instances Percentage of group reporting at least one next destination output Number of awards in group Number reporting at least one next destination output Total number of next destinations reported by the group Total number of unique next destinations reported by the group 45% 9,181 4,152 13,304 12,834 The distribution of next destination by sector for awards can be seen in Figure 8. By far the most frequently reported destination (67%), grouping all roles together, was to the academic/other research council sector. Figure 8: Next destination sector 9% Academic/other RC 19% Charitable/healthcare Other public sector (e.g. research agency/government) 1% 4% 67% Private (profit making e.g. industry, commerce) Unknown Table 12 compares the roles of staff employed on EPSRC grants with their roles at their next destination after leaving the EPSRC project (this, typically, is when the funding ends). Post doctoral researchers together with research students/researchers (NoPhD) account for over 90% of the staff reported as leaving and most of these remain in a research occupation. Note: The data reported above and in Table 12 cover some 12,000 next destination reports submitted to EPSRC by Principal Investigators of grants ending after 1 November The approach to collection of next destination information has changed significantly over the period since then, and hence should be considered broadly indicative only. 20

24 Table 12: Next destinations by role Role after group ( next destination ) Role in group Post doctoral researcher Research student/ researcher (no PhD) Research fellow/ research project leader Other research occcupation Post doctoral researcher 14% 0% 4% 26% Research student/researcher (no PhD) 7% 3% 1% 7% Technical support 0% 0% 0% 2% Research fellow/research project leader 0% 0% 1% 1% Total 21% 3% 6% 36% Note: Management/admin/policy and other have been removed from role in group. Full-time lecturer/teacher 12% 2% 0% 1% 14% Management/ admin/policy Other Total 1% 7% 66% 1% 7% 27% 1% 1% 4% 0% 1% 4% 3% 16% 100% 21

25 Influence on policy and practice Over 900 policy influences were reported, demonstrating the significant relevance of EPSRC funded research to across a wide range of sectors from energy and environment through communications, healthcare, and transport, to security and the creative economy. This section is intended to capture details of any significant impact on policy or practice that has been or is being realised as a result of the research, including impacts on areas of relevance to society such as housing, transport and schools. Table 13: Policy influence summary by instances Percentage of group reporting at least one policy influence Number of awards in group Number reporting at least one policy influence Total number of policy influences reported by the group Total number of unique policy influences reported by the group 5% 9, , For research grant holders, the most commonly cited form of influence was participation in advisory committees followed by membership of a guidance committee (41% combined). The full range of reported types of policy influence are shown in Figure 9. Figure 9: Type/method of influence on policy and practice (unique counts) Participation in an advisory committee Membership of a guidance committee Influenced training of practitioners or researchers Participation in a national consultation Gave evidence to a government review Citation in other policy documents Implementation circular/rapid advice/letter to e.g. Ministry of Health Not stated Citation in systematic reviews Citation in clinical reviews

26 Most reports of influence on policy and practice are at the national level, with more than twice as many instances reported at that level compared to the international level. Table 14: Geographical extent of policy influence Geographic reach National International Local/municipal/regional Not stated Total Each instance of influence can be cited as relevant to one or more sectors of national economic importance. The relative frequency with which instances of influence are reported as relevant to specific industry sectors is shown in Figure 10. The energy sector is cited more than any other, with almost 18% of all relevant sector citations. Figure 10: Sectors of relevance to reported instances of policy influence Energy Environment Digital/communication/information technologies Education Healthcare Transport Manufacturing, including industrial biotechnology Aerospace, defence and marine Electronics Government, democracy and justice Communities and social services/policy Construction Pharmaceuticals and medical biotechnology Culture, heritage, museums and collections Security and diplomacy Agriculture, food and drink Chemicals Other Creative economy Financial services, and management consultancy Leisure activities, including sports, recreation and tourism Retail % 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% Percentage of awards reporting policy influence (number to right of bar) 23

27 Physical outputs summary Significant numbers of physical outputs were reported: software and technical products (1,403), research tools and methods (652), research databases and models (685), artistic/creative outputs (487), products and interventions (90). In these sections researchers are asked to record details of new or significantly improved products and/or processes that have arisen directly from the work funded. They are reminded, however, not to disclose confidential information, for example about a new technology if the relevant intellectual property has not yet been fully protected. Five main outcome types are captured: artistic/creative; software/technical products; research tools and methods; research databases and models; products and interventions. Table 15 summarises how often these have been reported, and Figure 11 shows graphically the relative volume of unique reports of each type. Table 15: Physical outputs summary Outcome type % of total awards reporting at least one of this outcome type No. of awards reporting at least one of this outcome type Total reported by this outcome type Total reported by this outcome type (unique) Artistic/creative 1.8% Products and interventions 0.7% Research databases and models 4.9% Research tools and methods 4.2% Software/technical products 8.9% 816 1,599 1,403 Figure 11: Physical outputs - unique counts by type Software/technical products Research database and models Research tools and methods Artistic/creative Products and interventions ,000 1,200 1,400 1,600 number of outputs 24

28 Intellectual property and licensing More than 1,000 patent applications have arisen over the past decade from EPSRC-funded research; approximately one half have been granted and one third are still under examination. In the intellectual property and licensing section researchers are asked to report new intellectual property, to confirm if and how it has been protected, and to indicate if it has been licensed to others to exploit. Researchers are further asked to update (if appropriate) the status of any intellectual property records previously reported. Table 16: Intellectual property outputs summary Percentage of group reporting at least one IP Number of awards in group Number reporting at least one instance of IP Total number of instances of IP reported by the group Total number of unique instances of IP reported by the group 6% 9, , From Table 16 it can be seen that 856 unique instances of IP were reported by 6% of the 9,181 EPSRC awards; this is consistent with the fact that IP is only one of a number of routes to impact. Figure 12 shows that just over half of these instances are reported to have been licensed indicating a significant level of active exploitation of IP. Figure 12: Licensing of IP attributed to EPSRC grants Licensing arrangements are confidential Licence agreement has been reached Not licensed The data submitted via Researchfish in 2016 has added to the patent information held by EPSRC. It should be noted that: The status of a patent application can vary between under examination, granted and withdrawn/ terminated and can change at any time after it was filed; A patent application can be the first in a new family of patents related to the same underpinning invention or can be additional to an existing family of patents, for example by exploiting an original invention in a novel way. 25

29 To explore these aspects EPSRC works with patent specialists to clean the data and analyse the status and patent families of all applications which researchers have indicated arose from EPSRC-funded research. The most recent analysis was undertaken in June The cleaned data (see figures 13 and 14) shows that over the past decade more than 1,000 patent applications have arisen from EPSRC-funded research, and that of these approximately one half have been granted, one third are still under examination, and one fifth have been terminated or withdrawn. Furthermore, a quarter of all applications have initiated new patent families because these are not classified as extending or adding to an existing patent family they arguably represent more fundamentally new innovations. Figure 13: Patent applications arising from EPSRC grants, by June 2016 status and year of application Under examination Granted Withdrawn/terminated Year of application Count of patent applications Figure 14: Proportion of patent applications arising from EPSRC grants which have given rise to a new patent family, by year of application Year of application % 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% New patent families Extended patent families 26

30 Spin-outs 548 spin-outs created since EPSRC research stimulates innovation in a wide range of sectors electronics, healthcare and manufacturing have the most spin-outs. For the spin-outs section, researchers are asked to record details about new private organisations (for profit or not-for-profit) established as a result of their research and also existing private organisations (for profit or not-for-profit) where the strategy and/or turnover has developed directly as an output from their research award. In addition they are asked to report changes in the status of private sector organisations previously reported (for example if they have merged, expanded, been acquired or dissolved). Table 17: Spin-outs summary Percentage of group reporting at least one spin-out Number of awards in group Number reporting at least one spin-out Total number of spin-outs reported by the group Total number of unique spin-outs reported by the group 3% 9, Due to changes in the way this data has been gathered in recent years, and the fact that spin-out companies are often created well after the basic research is completed, the records in Researchfish under-report the actual number. Taking account of data from previous reporting systems and other sources such as the recent REF and Companies House, EPSRC is aware of a total of 548 spin-out companies created since 2004, and this more comprehensive overview is presented in the figures below while showing the contribution of the Researchfish data. It is expected that in future years the Researchfish records will represent the primary source of this information. Analysis of Companies House data shows that around 80% of the known spin-out companies are still active. Figure 15: Number of spin-out companies by year of incorporation, showing source of data Year of incorporation Number of companies Data sources: Researchfish Other 27

31 EPSRC record the relevance of research grant applications to different industry sectors, and the main areas of activity of known spin-out companies can be identified from analysis of SIC codes. These are presented in Figures 16 and 17. While grants most closely associated with the healthcare and electronics sectors have produced relatively more spin-out companies than other sectors, the data demonstrates that EPSRC research stimulates innovation and activity in a wide range of sectors. Figure 16: Proportion of spin-out companies attributed to grants, by relevance of grants to industry sectors Sectors to which grants were considered relevant Electronics Healthcare Manufacturing Energy Communications Information technologies Construction Pharmaceuticals and biotechnology Financial services Aerospace, defence and marine Transport systems and vehicles Chemicals Retail Creative industries Environment R&D Water Food and drink Education Technical consultancy Sports and recreation Broad relevance (basic underpinning research) 0% 5% 10% 15% Proportion of reported spin out companies attributed to grants Figure 17: Proportion of identified spin-out companies by standard industrial classification (SIC) Natural sciences, engineering Scientific research and development Computer programming, consultancy and related activities Other professional, scientific and technical activities Manufacture of computer, electronic and optical products Office business support activities Other manufacturing Architectural/engineering testing and analysis Other Publishing activities Telecommunications Human health activities Special purpose machinery Manufacture of chemical products Manufacture of electrical equipment Activities of head offices; management consultancy activities Information service activities Manufacture of basic pharmaceuticals Other personal service activities Biotechnology Social sciences/humanities 0% 5% 10% 15% 20% 25% 30% 35% 40% Proportion of identified spin out companies, by SIC Divsion Note: Some aggregation of categories has been undertaken to improve readability, and sub-activity is shown for companies in the SIC category scientific research & development. 28

32 Engagement activities Researchers reported engagement activities in 25% of EPSRC awards. The most popular method of engagement was a talk or presentation (47%), followed by participation in an activity, workshop or similar (26%). For this section researchers are asked to record details of any activity where aspects of the research supported by this grant were disseminated/communicated both to academic and non-academic audiences. Table 18 shows that 25% of awards reported at least one engagement activity. Table 18: Engagement activities summary Percentage of group reporting at least one engagement activity output Number of awards in group Number reporting at least one engagement activity output Total number of engagement activity outputs Total number of unique engagement activities reported by the group 25% 9,181 2,292 17,397 15,856 Under activity type, researchers are asked to note any activity intended to communicate the research beyond their normal peer group and which involved them or a member of their team. The option scientific meeting (conference/symposium etc.) was not available as a selection for the 2016 submission period it was felt that attending conferences is normal research activity and not a special instance of dissemination to be reported here. Guidance has been added to Researchfish indicating that a personal invitation as a keynote or other named speaker at a conference, should be recorded under awards and recognition and work included in published conference proceedings should be recorded under publications. Figure 18 shows the proportion of reported engagement activities by type and primary audience. The data shows that almost irrespective of the audience the commonest methods of engagement are through giving talks or presentations (47% of the total), and through participation in workshops or similar activities (26% of the total). The data also shows that most of the engagement activity (36%) was directed towards professional practitioners, (engineering professionals, for example) and just over 20% was delivered with the public as the primary audience. 29

33 Figure 18: Proportion of engagement activity reported by EPSRC grant holders, by type of activity and primary audience Research participants Media (as a channel to the public) Reported primary audience Schools Postgraduate students Policymakers Industry/business Other academic audiences Public Professional practitioners 0% 5% 10% 15% 20% 25% 30% 35% 40% Proportion of reported engagement activity A talk or presentation A formal working group, expert panel or dialogue A press release, press conference or response to a media enquiry/interview Engagement focused website, blog or social media channel Participation in an activity, workshop or similar A magazine, newsletter or online publication Participation in an open day or visit at my research institution A broadcast e.g. TV/radio/film/podcast (other than news/press) Researchers are asked to indicate the geographical reach of the engagement activity, for example whether the audience consisted of attendees from their immediate area or the wider local region, or whether it was a national or international event (Figure 19). Most (46.9%) recorded international engagement activity, followed by national (32.2%). Figure 19: Geographical reach of the reported engagement activity International 46.9% National 32.2% Local 10.4% Regional 10.5% 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 Number of outcomes (percentage after bar) 30

34 Awards and recognition EPSRC-funded research leads to career prestige in at least 20% of EPSRC awards, most frequently in the form of personal invitations to serve as conference key note speakers. For the awards and recognition section of Researchfish, researchers are asked to enter details of any significant award made to them or members of their team, in recognition of their funded research where the award was open to nominations at least on a regional level. Included also are invitations to conferences as a named/keynote speaker. Awards or appointments made below regional level (e.g. within the employing university) should not be reported. Table 19: Awards and recognition summary by instances Percentage of group reporting at least one recognition Number of awards in group Number reporting at least one recognition Total number of recognitions reported by the group Total number of unique recognitions reported by the group 20% 9,181 1,880 7,895 6,258 Figure 20 shows the distribution of awards and recognition by type. The type most frequently reported by awards was personally asked as a key note speaker to a conference (51.2% of this outcome type on awards) followed by research prize (16.7%). Figure 20: Awards and recognition by type Personally asked as a keynote speaker to a conference 51.2% Research prize Other 16.7% 15.1% Awarded honorary membership, or a fellowship, of a learned society Prestigious/honorary/advisory position to an external body Medal National honour e.g. Order of Chivalry, OBE Honorary degree 7.1% 5.5% 3.8% 0.4% 0.2% Number of outcomes (percentage after bar) It is worth noting that Other includes attracted visiting staff or user to your research group, appointed as the editor/advisor to a journal or book series, poster/abstract prize, NIHR senior investigator/clinical excellence award. 31

35 Researchers reporting in this category are asked to select the level at which nominations were open for this award. For example, nominations for Nobel prizes are open on an international level, Orders Of Chivalry (Knighthood or OBE) are national and prizes open to all researchers within a particular region are regional. Figure 21 shows that for awards, 69.5% are on an international level, 25.4% national and 5.2% regional and this distribution was the same in Figure 21: Reported level of recognition Continental/international 69.5% National (any country) 25.4% Regional (any country) 5.2% ,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 Number of outcomes (percentage after bar) 32

36 Secondments 14% of awards reported secondments. In this section, researchers are asked to record details of temporary secondments, placements and internships that have involved them or a member of their team whilst working on the project. These may be team members going elsewhere or colleagues coming into their team for a period of time. Table 20: Secondments summary Percentage of group reporting at least one secondment Number of awards in group Number reporting at least one secondment Total number of secondments reported by the group Total number of unique secondments reported by the group 14% 9,181 1,311 2,450 2,410 33

37 Students section Key findings and narrative impact Students reported key findings most frequently of relevance to energy, pharmaceuticals and aerospace, defence and marine sectors and narrative impact most often in aerospace, defence and marine, energy and digital/communications. Nearly half of students reporting impact type reported economic impact and significant levels of other impacts (societal, cultural, policy) were also reported. Of the 3,596 students who reported key findings in 2016, the most commonly identified sectors of relevance were: energy (23.2%), pharmaceuticals and medical biotechnology (23.1%) and aerospace, defence and marine (20.9%). 935 students reported narrative impact - aerospace, defence and marine (20%), energy (16%) and digital/communications (16%). The number of key findings and narrative impact student records were low - many students were still engaged in their studies at the time of reporting. Where key findings/narrative impact were submitted, students reported a wide range of sectors expected to benefit from their research (Figure 22). The sectors selected may reflect the focus of the Doctoral Training Centres (DTCs) as well as the main activities of the private sector organisations supporting the training. Figure 22: Sectors in which key findings were reported/impacts achieved (students) Aerospace, defence and marine Energy Environment Digital/communication/information technologies Manufacturing, including industrial biotechnology Pharmaceuticals and medical biotechnology Healthcare Chemicals Other Electronics Transport Education Culture, heritage, museums and collections Construction Agriculture, food and drink Government, democracy and justice Creative economy Security and diplomacy Communities and social services/policy Financial services, and management consultancy Leisure activities, including sports, recreation and tourism Retail 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% Percentage of studentships reporting Narrative impact Key findings Note: students were able to select multiple impact sectors. This is a different pattern from the main sectors reported in key findings and narrative impact by research grants which were energy, digital/ communications and healthcare. 34

38 Of the students on which narrative impact were reported in 2016, the most commonly identified impact type was economic, reported by 47.1% of students. Significant levels of other types of impact were also reported (see Figure 23). N.B. Students were able to select multiple types of impact. Figure 23: Types of impact reported in narrative impact Economic Policy & public services Societal Cultural number of students 35

39 Publications The most commonly reported outcome by EPSRC-supported students (over 8,000 publications) representing 23% of all their reported outputs. 38% of students reported at least one publication. Journal article was the commonest type of publication reported (55%). Publications are the most often reported research output accounting for over 23% of all outputs attributed by students to their EPSRC-supported research. Table 21: Publications summary by instances Publications summary Percentage of group reporting at least one publication Number of students in group Number reporting at least one publication Total number of publications reported by the group Total number of unique publications reported by the group Students 38% 7,869 2,985 8,068 8,061 38% of students reported at least one publication. Journal article was the commonest type of publication reported for students (55%), followed by conference papers (29%). Data provided by research organisations indicate that EPSRC support ended in or before 2015 for approximately 40% of students who reported publications, while 60% were still engaged being supported by EPSRC funds. Few of the students who submitted publications and whose EPSRC support ended before 2015, reported their PhD thesis as a publication (16%). One reason for this under-reporting is that it can take several months after the studentship has ended for the thesis to be ready for publication. Another is that it is hard to keep in contact with students who have finished their studies and left that organisation. The Researchfish guidance for students needs to state more clearly that students are expected to report their PhD theses under publications. Figure 24: Distribution of publication types 1% 2% Journal article 9% 4% Conference proceedings abstract Other Book/chapter/book edit 29% 55% Technical report/standard Thesis/other student publication Consultancy/policy report 36

40 Collaborations and partnerships 12% of awards reported collaborations, over half of which were new collaborations (8% of awards). Approximately one quarter (25%) of collaborations involved private sector organisations and 12% public sector. Most frequently reported international collaborator: USA. Table 22: Collaborations summary by instances Publications summary Percentage of group reporting at least one collaboration Number of students in group Number reporting at least one collaboration Total number of collaborations reported by the group Total number of unique collaborations reported by the group All 12% 7, ,669 1,662 New collaborations 8% 616 1,055 1,054 Students reported collaboration outcomes in 12% of those submitting to Researchfish in For new collaborations, where the year the collaboration started was after the award start date, outcomes were reported in 8% of students. Overall, the majority of the collaborators were from the academic sector (55%) followed by the private sector (25%) and public sector (12%). The location of the collaborators selected most frequently was United Kingdom (59%) followed by EU (18.5%) and USA (10.2%). A list of the top private sector collaborators reported in Researchfish by students can be seen in Table 23. Table 23: Top private sector collaborators by number Collaborator Rolls Royce Group Plc AstraZeneca GlaxoSmithKline (GSK) Johnson Matthey Atomic Weapons Establishment (AWE) Pfizer Ltd BASF Arup Group Procter & Gamble BAE Systems Procter & Gamble Technical Centres Ltd Thales Group Tata Steel Europe 37

41 The UK private sector collaborations were mapped to industrial sector by Standard Industrial Classification (SIC) code and this showed that most (34%) were in manufacturing followed by professional, scientific and technical (26.5%). Table 24: Industrial sector of private sector collaborations Industry Count % Manufacturing Professional, scientific and technical Wholesale and retail trade, repair of motor vehicles and motorcycles Administrative and support services Information and communication Water supply, sewerage and waste management Construction Human health and social work Education Financial and insurance activities Transportation and storage Mining and quarrying Other Total Footnote to Table 24: Other includes: electricity, gas, steam and air conditioning supply, accommodation and food services, public administration and defence, agriculture, forestry and fishing, arts, entertainment and recreation, real estate activities, other services. The 117 collaborations from the manufacturing sector were broken down further into subtypes. It was found that 36% were from the manufacture of computer, electronic and light electrical goods and 25% were from the manufacture coke, petroleum, chemicals, and pharmaceuticals. 38

42 Facilities 355 EPSRC-funded research students reported having accessed almost 190 different facilities during their training. A high proportion (almost 70%) of these facilities are in the UK, with 22% elsewhere in Europe, 5% in the USA and 4% in other countries. Table 25: Use of facilities summary by instances Percentage of group reporting at least one instance of facility usage Number of students in group Number reporting at least one instance of facility usage Total number of instances of facilities usage reported by the group Total number of unique instances of facilities usage reported by the group 5% 7, EPSRC-funded research students reported having accessed almost 190 different facilities during their training. A high proportion (almost 70%) of these facilities are in the UK, with 22% elsewhere in Europe, 5% in the USA and 4% in other countries. Table 26: Location and number of facilities used, and volume of grants reporting facility use Location UK Rest of EU USA Rest of World Total Count of facilities Unique reports of use by students The figures in Table 27 are the numbers of students which reported use of the named facilities. The figures are indicative rather than absolute because summarising the data is difficult due to lack of adoption of standardised names and the frequent use of a site location name as a proxy for the actual facility used. 39

43 Table 27: Top 10 facilities most frequently used to support EPSRC-funded students Facility DIAMOND Synchrotron STFC ISIS Pulsed Neutron and Muon Source Other HPC Facility ARCHER/HECToR European Synchrotron Radiation Facility (ESRF) STFC Central Laser Facility Institut Laue-Langevin Neutron Scattering Facility Swiss Light Source (SLS)/Paul Scherrer Institute (PSI) 850MHz Solid State NMR Facility at Warwick EPSRC National Service for Computational Chemistry Software Research students

44 Next destination and skills 60% of students who reported their next destination remained in academic research organisations and 28% took up a role in the private sector. Only 2% of students reported in this category as this section is not relevant to students who are still studying. (see Table 28). Table 28: Next destinations summary by instances Percentage of group reporting at least one next destination output Number of students in group Number reporting at least one next destination output Total number of next destinations reported by the group Total number of unique next destinations reported by the group 2% 7, The distribution of next destination by sector for students can be seen in Figure 25. By far the most frequently reported destination (54%), grouping all roles together, was to the academic sector. The destination of leavers in higher education (DLHE) data shows a lower percentage of students going on to the academic sector (35%) and a higher percentage to the private sector (40%) and is likely to be a more accurate picture. The Researchfish sample of students is very small (approximately 200, see Table 29) and is considered likely to be skewed in favour of an academic next destination as this set are more likely to have been contacted by and have access to Researchfish. Figure 25: Next destination sector (role after group) 8% Academic/other 28% 54% Charitable/healthcare Other public sector (e.g. research agency/government) Private (profit making e.g. industry, commerce) 7% Unknown 3% The next destination role of most of the research students is to post doctoral researcher (45%), another research occupation (16%) or Other - including: engaged in study, health care/medical staff, unknown, not employed/maternity and other (25%). 41

45 Physical outputs summary Significant numbers of physical outputs were reported: research databases and models (424), software and technical products (330), research tools and methods (220), artistic/creative outputs (128), products and interventions (24). Table 29 summarises the outcomes from five of the outcome types: software/technical products, research tools and methods, research databases and models, artistic/creative and products and interventions. Table 29: Physical outputs summary Outcome type % of total students reporting at least one of this outcome type No. of students reporting at least one of this outcome type Total reported by this outcome type Total reported by this outcome type (unique) Artistic/creative 1.0% Products and interventions 0.3% Research databases and models 3.6% Research tools and methods 2.2% Software/technical products 3.1% For students, research database and models had the most outputs recorded out of these five outcome types, followed by software/technical products and research tools and methods. Awards recorded the most outputs in software/technical products followed by research database and models and then research tools and methods. Figure 26: Physical outputs summary Research database and models Software/technical products Research tools and methods Artistic/creative Products and interventions Number of outputs 42

46 Intellectual property and licensing Students reported more than 50 patent applications and most of these (80%) are published. For the intellectual property and licensing section we asked students to report new intellectual property, to confirm if and how it has been protected, and to indicate if it has been licensed to others to exploit. Table 30 shows that instances of IP were reported from 1% of the 7,869 students. Table 30: Intellectual property outputs summary Percentage of group reporting at least one IP Number of students in group Number reporting at least one instance of IP Total number of instances of IP reported by the group Total number of unique instances of IP reported by the group 1% 7, Most instances (80%) of IP reported by students are published patent applications, however it is notable that some refer to already granted patents in which the student is a named inventor. Licence agreements have been reached in 13% of instances of IP reported by students with a further 21% described as commercial in confidence 43

47 Spin-outs Students reported 14 spin-outs created from the following sectors: healthcare, energy, communications, IT, creative industries, aerospace, defence and marine. 12 of these were only reported by students. Scoop Analytics Ltd is an example of a spin-out built on top of research derived from an EPSRC-funded PhD studentship at the school of computing science at the University of Glasgow. Established in 2015, Scoop monitors social media to give journalists and financial organisations instant alerts to breaking news stories as they happen and before they hit mainstream media. 44

48 Engagement activities 18% of students reported engagement activities. The most popular method of engagement was a talk or presentation (49%), followed by participation in an activity, workshop or similar (35%). Table 31 shows that 18% of students reported at least one engagement activity. Table 31: Engagement activities summary Percentage of group reporting at least one engagement activity output Number of students in group Number reporting at least one engagement activity output Total number of engagement activity outputs Total number of unique engagement activities reported by the group 18% 7,869 1,379 3,898 3,878 Figure 27 shows the number of engagement activities by type. The most popular method of engagement for students was a talk or presentation (49.1% of the student total for this outcome type), followed by participation in an activity, workshop or similar (35.1%). Figure 27: Engagement activity by type A talk or presentation 49.1% Participation in an activity, workshop or similar 35.1% Participation in an open day or visit at my research institution 6.9% A formal working group, expert panel or dialogue 3.9% A magazine, newsletter or online publication 1.8% Engagement focused website, blog or social media channel 1.3% A press release, press conference or response to a media enquiry/interview 1.0% A broadcast e.g. TV/radio/film/podcast (other than news/press) 0.8% ,200 1,600 2,000 Number of outcomes (percentage after bar) Figure 28 gives a breakdown of the engagement audience. For students, most of the dissemination activity was aimed at other postgraduate students (23.6%), public/other audiences (23.2%) and professional practitioners (21.8%). Public and schools together account for nearly 40% of the student dissemination engagement audience. This shows that students are making a very positive contribution to public engagement through outreach events etc. 45

49 Figure 28: Proportions of engagement activity reported by EPSRC students, by type of activity and primary audience Research participants Other academic audiences Media (as a channel to the public) Policymakers Other Industry/business Schools Professional practitioners Public Postgraduate students 0% 5% 10% 15% 20% 25% A talk or presentation Participation in an activity, workshop or similar A formal working group, expert panel or dialogue A magazine, newsletter or online publication A press release, press conference or response to a media enquiry/interview Participation in an open day or visit at my research institution Engagement focused website, blog or social media channel A broadcast e.g. TV/radio/film/podcast (other than news/press) Footnote to Figure 28: Audience type Other includes undergraduate students, study participants or study members, third sector organisations, patients, carers and/or patient groups, supporters. Note: Unknown audience has been removed from Figure 28. Students are asked to indicate the geographical reach of the engagement activity, for example whether the audience consisted of attendees from their immediate area or the wider local region, or whether it was a national or international event (Figure 29). Most (38.3%) recorded international engagement activity, followed by national (23.9%). Figure 29: Geographical reach of the reported engagement activity International 38.3% National 23.9% Local 21.1% Regional 16.7% ,000 1,200 1,400 1,600 46

50 Secondments 13% of students reported secondments. In this section, students are asked to record details of temporary secondments, placements and internships that have involved them whilst working on the project. Table 32: Secondments summary Percentage of group reporting at least one secondment Number of students in group Number reporting at least one secondment Total number of secondments reported by the group Total number of unique secondments reported by the group 13% 7,869 1,047 1,404 1,242 47

51 Case studies Reported in Researchfish, the following show a range of impacts arising from EPSRC-supported research, including the creation of spin-out companies, collaborations with industry and public sector organisations, acceleration of new technologies and products to the market place and support for the talented scientists, engineers and postgraduate research students that are helping to drive UK innovation. Spin-out company worth 130m A company founded to commercialise artificial intelligence technology developed at the University of Aberdeen is just one contribution made by EPSRC-funded researchers to the big data revolution. Data2Text was formed in 2010 to take forward Natural Language Generation (NLG) research. NLG is a form of artificial intelligence developed to communicate information extracted from complex data sources in natural language, automatically generating written reports which would take expert analysts hours to complete. The company s goal was to build articulate machines which communicate with people in the same way that other people do. The technology developed is being used by the Met Office to generate high quality text for weather forecasts. A global oil and gas company is also using it to process vast amounts of data from production platforms in deep-water Gulf of Mexico and convert it in real time to the language that an operator or engineer speaks. Data2Text was acquired for a combination of equity and cash by software development business Arria. The resulting company, Arria NLG, floated in early December 2013 on the Alternative Investment Market under the symbol NLG, and was valued at 130m. Pay-as-you-go technology brings power to off-grid markets Solar energy spin-out company Eight19 Ltd was established in 2010 from the EPSRCfunded Cambridge Innovation and Knowledge Centre (Advanced Manufacturing Technologies for Photonics and Electronics), to build on the development of clean technology to enable a new generation of low-cost, flexible plastic solar cells that have the potential to dramatically reduce the manufacturing cost and increase the throughput of solar technology. The company s Indigo pay-as-you-go, low cost, solar technology product for the developing world, won the World Business and Development Award at the RIO+20 UN Climate Conference in 2012 and is currently being used in Kenya, Malawi, Zambia, South Sudan, Uganda and South Africa, providing basic needs that are regarded as routine in developed countries. Eight19 has gone on to set up a new company to further develop the Indigo technology. Azuri Technologies Ltd has raised capital from IP Group plc to scale-up its deployment of the technology, bringing power to off-grid customers worldwide and allowing users to light their homes and recharge mobile phones as a service paid for using scratchcards. Currently 1.6bn people, over one fifth of the world s population, lack access to electricity via a grid and pay high prices for fuels such as kerosene to serve their basic needs. 48

52 Designs for improving hospital patient care A three-year multidisciplinary EPSRC-funded project with research teams from the Royal College of Art (RCA) and Imperial College London (ICL), was set up with the aim to reduce medical error by creating a better fit between healthcare processes on surgical wards and the equipment and products that support them. (Healthcare Innovation Exchange). A team of designers, researchers and clinicians embedded in London s St Mary s Hospital are utilising many aspects of the DOME project. The Designing Out Medical Error (DOME) research team mapped surgical processes with NHS staff and patients; investigated how safety is managed; and used novel research techniques to identify and prioritise the five most error-prone processes on surgical wards - hand washing, information handover, vital signs monitoring, isolation of infection and medication delivery. Interventions were designed for each process and tested in a simulated ward environment. These included the CareCentre, an all-in-one unit for the equipment needed for patient care in the bed space, now commercially available from UK manufacturer and DOME partner Bristol Maid; a communication campaign for hand hygiene; and a new easy to clean and use trolley to monitor vital signs being developed with USA manufacturer Humanscale. DOME served as a forerunner to a new RCA/ICL design-led innovation centre, the HELIX Centre Copyright: RCA Award winning animation studios benefit from working with doctoral students Based at the Universities of Bath and Bournemouth, the EPSRC Centre for Doctoral Training in Digital Entertainment funds doctoral students in digital games, visual effects and animation. Working with a network of small and median enterprises, the students gain exceptional industry experience and companies benefit from access to highly-skilled industry-prepared students. One business which has benefited from this relationship is WONKY, an award-winning animation studio with clients including the BBC, British Council and UNICEF. Working with a student from the centre, the company has been able to strengthen its positioning in digital environments and has an increased online presence through the digital delivery of short films and social media networks. Another student has helped Aardman Animations to gain insight into conducting large in-house research projects supporting the development of new project areas. One of Europe s largest provider of visual effects for the film industry, Double Negative, has also benefited from new proven techniques for solving significant problems in the production process. 49

53 Standardisation delivers improved emergency service Ten years of research on ambulance and equipment design resulted in improved patient care and significant cost savings. EPSRC-funded research at Loughborough University prompted a national specification to standardise the design of emergency ambulances, with six of the 11 NHS Ambulance Trusts in England purchasing new single specification ambulances. The new specification produced national savings in excess of 2.5m in over three years and influenced international design. delivery of care without transportation to hospital and the findings have been used by the NHS Supply Chain in its purchase of replacement portable equipment. Before the standardisation, NHS Ambulance Trusts produced their own vehicle specification resulting in over 40 different designs of emergency ambulances in the UK. This presented an increased risk to patient safety as the location of equipment and consumables and interior layout varied in each vehicle. In response to the Department of Health s 2005 vision for future ambulance services, the research team looked at the safety of both patients and staff during the use of mobility equipment, as well as a range of issues relating to ambulance design including layout, securing people and equipment in transit, hygiene and communication. Other research looked into the development of new paramedic equipment and vehicles to support the Research expertise behind inward investment and new business EPSRC-funded internationally-leading Grid and Cloud computing research at Newcastle University - which is continuing under the EPSRC-led RCUKfunded Social Inclusion through the Digital Economy (SiDE) research hub - was behind a world-leader in open source software investing in the UK. The proximity to research expertise played a large part in the $10bn US-based Red Hat s decision to create a research and development centre at Newcastle University. The centre is Red Hat s only R&D centre outside of the USA and it is estimated to have contributed $15m GVA and 75 person-years of high-tech employment to the North East since strengthening the UK s position in the global marketplace with research tied to real product development and market demands. In collaboration with Newcastle City Council, SiDE has established the Cloud Innovation Centre. The centre facilitates collaboration between industry, the public sector, academic research and teaching, encouraging innovation and new business creation through research and development, knowledge transfer and skills improvement across the region. 50

54 The generator that transforms lives An affordable cooking stove that generates electricity could transform lives in the world s poorest communities. Developed by the University of Nottingham, the EPSRC supported Score project, the concept behind the Score-Stove (Stove for Cooking, Refrigeration and Electricity) is a novel application of thermo-acoustics, where heat is produced, converted into sound energy and then electricity. The stove can be fuelled by burning a range of biomass products such as wood or dung and converts the heat produced into electricity - enough to power lights and simultaneously charge a mobile phone. Any surplus electricity produced by households could be sold to neighbours. An early version Score-Stove primarily for use as a high-efficiency, low-smoke cooking stove, is being used by communities in Nepal, Kenya and Zambia. With the further development of the Score- Stove, the project is continuing to work with poor communities to develop local businesses for the manufacture, repair and innovative applications of Score. Aiming to improve the life of over 60m people in developing countries, the project began as a 2m research consortium. Its goal was to significantly improve health, quality of life, economic growth and social and educational opportunities, thus reducing poverty in sub-saharan Africa, the Indian sub-continent and South America, by understanding the energy needs of their rural communities and working with them to develop the capability to manufacture an affordable versatile domestic appliance. 51