EU Support for Innovation and Market Uptake in Smart Buildings under the Horizon 2020 Framework Programme

Abstract: At the end of November 2016, the European Commission tabled the Clean Energy for All Europeans package, which represents a large set of proposals for several key directives related to energy. The package included proposed revisions to the Energy Performance of Buildings Directive (EPBD) which seek to update and streamline the Directive in several areas, including provisions to ensure buildings operate efficiently by encouraging the uptake of Information and Communication Technologies (ICT) and smart technologies. Although it can be argued that there is at present no commonly accepted definition of a “smart building”, the Commission’s proposed revision refers to three key features of a possible indicator of “smartness” in buildings: the technological readiness of a building to (1) interact with its occupants; (2) to interact with the grid; and (3) to manage itself efficiently. Using these three pillars of “smartness” as a methodological starting point, this paper identifies and analyses recent and ongoing Horizon 2020 research, innovation and market uptake projects which are investigating “smart buildings”. The research maps and examines the tasks, scope and innovations in areas that include building automation and control systems, demand response, energy management, ICT and user interfaces for energy efficiency.


Introduction: Purpose of the Research
This paper seeks to answer the following question: "How is Europe's Horizon 2020 framework programme supporting research, innovation and market uptake of smart buildings?"This apparently straightforward question is in fact relatively difficult to answer, partly because the term "smart buildings" is a frequently used but ill-defined [1] term that cuts across several themes including ICT (Information and Communication Technologies), energy efficiency, renewable energy technologies, design and construction techniques, and the dynamic interaction of a building with its occupants.A simple answer to the question might be to express the amount of grant funding allocated to projects working on smart buildings, or to count the overall number of projects supported.However, statistics such as these would in practice be relatively difficult to obtain, for several reasons.First, it would require projects across Horizon 2020 to apply a common and precisely defined understanding of the term "smart building", with the added complication that in this context we are only concerned with energy efficiency-related smartness.Second, it would depend on databases of projects being searchable using this commonly agreed indicator.The third reason why statistics would not be easy to come by is that one aspect of a smart building might be investigated in a particular task or work package within a larger project, which might not identify itself (for the purposes of a database search) as a "smart buildings" project.The research problem to be tackled in this paper is therefore to define the Buildings 2017, 7, 105; doi:10.3390/buildings7040105www.mdpi.com/journal/buildings"smart building", to search for and find projects that match this definition across the Horizon 2020 programme, and to analyse those results in a mapping exercise.
The overall purpose of this research is to generate new data that will be useful for both researchers and policymakers.Horizon 2020 projects can sometimes be large and complex, and it can be difficult for persons outside of the project consortium itself to appreciate the precise scope of work and the nature of individual tasks, from publicly available information such as abstracts.This is especially the case when projects are ongoing and results are yet to be published.Potentially useful data such as lists of deliverables and detailed breakdowns of tasks are not normally publicly available.Because of this, it can be difficult to establish the scale and extent of research efforts being made across the programme on any particular subject, beyond relatively simplistic estimations.Therefore in order to discern the extent of research and funding efforts in smart buildings it is necessary to generate new data, analyse it and place it in the public domain.This kind of mapping should offer a more precise picture, enabling better understanding of the status quo, more effective prioritisation of future research needs, and potentially help align EU funded research and innovation with national and private sector efforts.

Policy Context
The European Union (EU) has long pursued strategies to improve the energy performance of buildings, and these have been steadily increasing in ambition and scope.The Energy Performance of Buildings Directive (EPBD) [2] was published in 2002 and followed earlier EU legislation in the previous decade on energy efficiency measures in buildings [3] and their technical systems [4].In March 2007, EU Heads of State and Government agreed on Europe's 2020 Strategy for smart, sustainable and inclusive growth [5].This strategy set EU-wide targets on greenhouse gas emissions (a reduction of at least 20% by 2020, from a 1990 baseline), energy efficiency (savings of 20% by 2020, expressed in absolute) and uptake of renewable energy (achievement of a 20% share of renewable energy in EU energy consumption by 2020).In the same year, the European Commission also published the Strategic Energy Technology (SET) Plan [6] which aimed to coordinate EU, national and private funding efforts in technology research and development in order to accelerate the EU's transformation to a low-carbon energy system [7].In this context, the European Commission proposed in 2008 a recast of the EPBD which was adopted by the European Parliament and Council in May 2010 [8].The European Commission stepped up its efforts in January [9] and July [10] 2014 by proposing more ambitious energy efficiency targets by 2030, which were adopted by the European Council in October of that year [11]: a 40% reduction in greenhouse gas emissions from the 1990 baseline, a 27% share of renewable energy consumption, and a 27% improvement in energy efficiency, which would be reviewed by 2020 having in mind an EU level of 30%.In February 2015, the Commission adopted the Communication on the European Energy Union [12] which highlights energy efficiency, especially in buildings, as well as research and innovation as key "dimensions" that would drive Europe through an energy transition leading to greater energy security, sustainability and competitiveness [12].
More recently, in November 2016 the European Commission published a large package of legislative proposals under the umbrella title "Clean Energy for all Europeans" [13].Taken together, the proposals would adapt several directives relating to energy efficiency, energy security, renewable energy, electricity market design, and energy governance.Among the Directives put forward for review was the EPBD, with proposals to encourage the use of ICT and smart technologies to ensure that buildings operate as efficiently as possible [14].For the first time, European legislation was being squarely aimed at encouraging the uptake of "smart buildings".Specifically, the EPBD proposal opened the possibility of using building automation and control systems as an alternative to physical inspections of technical buildings systems [14].It also sought to facilitate the uptake of electric vehicles by ensuring that buildings would be ready to connect with them [14] and it introduced an indicator that would assess the degree to which a building can be deemed 'smart' [14].This 'Smart Readiness Indicator' would enable assessment of the "technological readiness" of buildings to manage their own energy performance, to interact with users and with the wider grid [14].

The Horizon 2020 Programme and Buildings' Energy Performance
In parallel with policy developments, the European Commission has over a number of years supported research, innovation and market uptake projects that help Europe use energy more sustainably [6].The current Horizon 2020 framework programme (2014-2020) [15] brings together research and innovation in an effort to ensure that scientific and technological breakthroughs lead to innovative products and services that tackle "the urgent challenges society faces" [15].These "Societal Challenges" sit alongside "Industrial Leadership" and "Excellent Science" as the three overarching priorities of Horizon 2020 [15].First, the "Excellent Science" priority includes support to scientists via the European Research Council; funding for specific "Future and Emerging Technologies"; and provisions for researchers to develop their careers through the Marie Skłodowska-Curie actions [15].Secondly, the "Industrial Leadership" priority aims to "build leadership" in specific "enabling and industrial technologies" as well as providing support for innovation in SME's and facilitating access to risk finance [15].Thirdly, the policy priorities of the Europe 2020 strategy [16] are reflected in a challenge-based approach to bring research results to the market according to "Societal Challenges", one of which is "Secure, clean and efficient energy" [16].This particular Societal Challenge aims to "support the transition to a reliable, sustainable and competitive energy system" [17].In the Work Programme this has been expressed in three priority "focus areas" for research, innovation and market uptake, namely "Energy Efficiency", "Competitive Low-Carbon Energy" and "Smart Cities and Communities" [18].Calls for proposals to receive funding are described in the work programme, and each call contains a number of specific topics [18].Work Programmes for the Societal Challenge "Secure, Clean and Efficient Energy" have been published for two-year periods, hence there is a Work Programme for 2014-2015 [18] and a separate Work Programme for 2016-2017 [19].
In practice, support is mobilised via several different types of funding schemes (grants for research and innovation; training and mobility; co-funding including public-private partnerships; grants to public procurement of innovation; support grants; debt finance and equity investments; prizes; and procurement) [20].The specific programme [21] for implementation of Horizon 2020 was approved by the Council in 2013.The Horizon 2020 programme has now reached its halfway point and the European Commission published an interim evaluation of its results at the end of May 2017 [22].This research paper therefore comes at an opportune time, although it does not form part of the Horizon 2020 Interim Evaluation, and the timing of this research was not influenced by the evaluation.
One particular development that coincided with the Horizon 2020 programme was the launch of the contractual Public-Private Partnerships for research (cPPP) in December 2013 [23].These partnerships aim to leverage private funds alongside EU grant funding [23] and are based on roadmaps that were developed together with industry stakeholders in an open consultation process [23].In total there are ten Public Private Partnerships [24], including the Energy-efficient Buildings Public Private Partnership (EeB PPP) which under Horizon 2020 continues activities that it began in 2009 [23].In 2010, the European Commission, through the EeB PPP, published a multi-annual roadmap and strategy for the years 2011-2013 [25] and it followed this up in 2013 with an updated roadmap for the years 2014-2020 [26].

Methodology
One advantage of having a single research and innovation programme for Europe is that this also offers the possibility to search and compare projects across all thematic priority areas in a single database.For smart buildings this is an important point, since the subject could be expected to involve projects funded under a variety of topics and types of action.The principal publicly accessible source of data on Horizon 2020 projects is the CORDIS database (Community Research and Development Information Service) [27].It contains data on signed grants and beneficiaries, abstracts, and certain publishable reports produced by projects [27].For individual projects, data is given on the names and addresses of the coordinator's organisation and any other project partners, and the EU funds they have received, the call topic, and the start and end dates [27].CORDIS enables searches to be carried out using keywords or free text, or by searching for project acronyms and reference numbers, or by topic, by type of action, or by a number of other criteria [27].It can also find projects that were funded under the same call for proposals as those found in an initial set of search results [27].For staff of the European Commission there is also a closed internal database called CORDA (COmmon Research DAtawarehouse), which is the European Framework Programmes' central repository of data [28].CORDA contains data concerning signed grants and beneficiaries and enables the preparation of statistical overview data with a degree of sorting according to filters such as thematic priorities, activities and countries [28].
In order to search effectively for projects, or for tasks within projects that are relevant to the development of smart buildings, one first needs to define what is understood by the term "smart building".Arguably, there is at present no single commonly accepted definition of this term [1].Various interpretations exist, including the term "Intelligent Building", which itself can be understood differently in different contexts [1].In its proposals for a revision of the EPBD [14], the European Commission refers to three key features of a possible indicator of "smartness" in buildings: the technological readiness of a building to (1) manage itself efficiently; (2) interact with its users; and (3) interact with the wider energy environment.For the purposes of this research, these three features are taken as a methodological starting point.Within these three features, further refinement and sub-division is necessary in order to form a practical basis for a comparative search and mapping exercise.A total of 16 features of energy-related smartness in buildings have therefore been set out and explained here below.

•
Full automation-the automatic centralised control of a building's energy-consuming technical building systems.

•
User interface-the interface by which the user can view, understand and control the building's energy consumption.This can be located either within the building or it can be remote (e.g., a hand-held device).

•
Control of entire building-the ability to control all of the energy-consuming systems and appliances in the entire building, including systems that might exert an influence over the building's energy consumption such as motorised windows and shading devices.

•
Control of individual appliances-the user's ability to control individual energy-consuming appliances within the building, other than via the appliance itself.

•
Implicit Demand Response-the ability of consumers to be "exposed to time-varying electricity prices or time-varying network tariffs (or both) that partly reflect the value or cost of electricity and/or transportation in different time periods and react to those price differences depending on their own possibilities" [29].

The Building's Ability to Manage Its Own Energy Consumption
• On-site storage-the storage of energy within a building's plot boundary.

•
Heating and cooling-the management of a building's heating and cooling systems.

•
Lighting-the management of a building's lighting systems.

•
Domestic Hot Water-the management of a building's domestic hot water systems.

•
Domestic appliances-the management of energy-consuming appliances situated within the building.

•
Self-learning and Artificial Intelligence-the ability of a building energy management system to apply decisions that influence the building's energy consumption without being explicitly programmed to do so.
• Optimization-the building energy management system's ability to optimally balance the building's energy loads including for generation, storage or consumption.

•
Electro-mobility and smart charging-the interaction on-site electric vehicles with the building's energy system.

•
Data privacy and protection-measures to ensure secure storage, management and use of data that is generated or used by technical building systems and appliances.

•
Explicit Demand Response-"the control of aggregated changes in load traded in electricity markets, providing a comparable resource to generation, and receiving comparable prices" [30].
In reality, although the 16 features are grouped under three principal headings, the divisions between them may not always be clear-cut and a degree of overlap can be expected.For example, self-learning or Artificial Intelligence could be expected to apply to a building's interaction with its users just as much as to the building's ability to manage itself.Nevertheless, for reasons of practicality these 16 features are felt to be the most appropriate to map the tasks that Horizon 2020 projects might be carrying out.The mapping exercise has therefore taken the form of manual searches, primarily using the CORDIS public database [27], for projects that are carrying out research, innovation and market uptake activities according to this list of 16 features that are relevant to smart buildings.The result has been a table matrix (Appendix A) which identifies the projects and their respective tasks corresponding to each of the 16 features.A simple list of the projects that were found to be relevant is also presented in Appendix B.
Three overarching criteria have also been applied to filter out irrelevant search results: 1.
Projects must be funded under the Horizon 2020 programme.This is to exclude other programmes such as FP7 which also appear in CORDIS [30].Although similar work may be ongoing under those programmes, this research covers the Horizon 2020 programme only.

2.
Project tasks must relate to the building level.This excludes, for example, tasks which relate to smart power grids at the district or city level but which are not directly related to the ways in which they might interact with buildings.

3.
Project tasks must relate to the sustainable use of energy in buildings.For example, some projects might include tasks relating to the Internet of Things and connected devices, but these would only be included in the study if those devices were energy-consuming appliances or technical systems that related to the building's energy use.
In addition to manual searches, some cross-referencing has been carried out.First, the outputs of the Horizon 2020 project EEBERS have been analysed [31], specifically the "Energy Efficient Buildings Projects Map" which aims to identify synergies between ICT related research and development in energy efficient buildings [32].Secondly, it has been possible to take advantage of unpublished research work by EASME staff members who had used the CORDA database to identify, for internal purposes, energy-related projects funded under the Horizon 2020 SME Instrument.These are typically two-stage projects where the first stage is a short study that might lead, depending on success with the second stage application, to a more detailed phase of real innovation activity.Many SME Instrument Phase I projects never go further than the first stage feasibility study.For this reason, only Phase II projects from the SME Instrument have been included in this research.

Searching and Mapping of Projects: Results
The database searches were carried out on several occasions during April and early May 2017, and included all Horizon 2020 actions that were either ongoing or completed by that date, since the start of the programme in 2014.Within the CORDIS database, the "Advanced Search" facility was used, which includes an ability to look for "Projects only" and exclude other potential results such as individual reports, news items and events [30].Filters were applied to restrict the "Programme" to Horizon 2020 only [30].Search terms were inserted into the "Search Terms" box and all other filters were kept open in order to avoid unwittingly restricting the results.The actual search terms used were "Smart Buildings", "Building Energy Management", "Energy Storage", "Demand Response", and "Optimization".Following these searches, further searches were deemed unnecessary as the results did not return any projects that had not already appeared.In addition to manual searches, CORDIS enables identification of all projects funded under the same Call as a single search result-hence, which also facilitates manual cross-referencing [30].This final act of cross-referencing was carried out in early July 2017 and resulted in modification of the results: two projects that had been funded under Smart Cities and Communities calls, which had been missed, were added.
The initial results produced a list of 65 Horizon 2020 projects, funded across a variety of thematic priorities, which might be carrying out relevant tasks based on the limited project information that is available from CORDIS [30].In order to verify and refine these results it was also necessary to analyse the individual project tasks, which required a level of information that is not available using CORDIS alone.CORDIS is a useful tool to gain a broad overview including project abstracts, overall budgets and publishable summary reports, but it does not at present allow detailed scrutiny of work plans down to the level of specific tasks and deliverables [30].The project websites were also analysed where these were available, but although they usually contain greater detail than the CORDIS entries, in most cases they still contain insufficiently detailed information for such an exercise.In any case, Horizon 2020 project websites are rarely online from the very beginning of a project, therefore actions which have only recently been funded can appear in CORDIS without yet having a project website in place.It was therefore necessary to identify and contact the individual European Commission staff members who were responsible for overseeing each project, and to make use of their in-depth knowledge to identify relevant project tasks.This was made possible by the European Commission's internal project management tools.In many cases, the project coordinators were also contacted after having obtained the permission of the relevant officer.Where the relevant officer or coordinator was unavailable, a detailed analysis of project tasks was carried out against the criteria set out in Section 3 above, using the Commission's internal tools which also offer full access to project documentation.As a result of this exercise, the initial list of 65 projects was refined and about one third were excluded as being out of scope.
The final results of the research have found 42 relevant Horizon 2020 actions.These are funded across 28 topics, the majority of which are Research and Innovation Actions (RIA) and Innovation Actions (IA).Four projects have been funded under the SME Instrument Phase 2. The search also revealed one action for research and innovation staff exchange (RISE) under the Marie Skłodowska Curie action, one European Research Council (ERC) proof of concept grant, and one Coordination and Support Action (CSA) funded under the Energy Efficiency call of 2015.The total budget costs for these 42 actions add up to 367.9 million Euros, of which the EU grant contribution is 304.1 million Euros.These statistics are presented here below in Table 1 and Figure 1.

Comparative Analysis
Figure 1 shows the number of projects supported by the different funding schemes while Figure 2 shows their total budget costs and the corresponding EU grant contribution.The majority of the projects are Innovation Actions (IA) and Research and Innovation Actions (RIA).However, when one looks at the allocation of funds, there is a large dominance of Innovation Actions accounting for some 82% of the EU grant contributions.This can be explained by the fact that the Innovation Actions include several projects funded under the Smart Cities and Communities calls, and these tend to have far larger budgets than other Horizon 2020 projects.Likewise, the almost negligible proportion of funds allocated to CSA, MSCA and ERC actions relates to the inherently smaller size of this kind of project.
The 42 projects have been funded across 28 topics within 23 calls for proposals from the years 2014, 2015 and 2016 (see Figure 3).Of the 42 projects, 15 have been funded under the Energy Efficiency calls; seven are from the Smart Cities and Communities calls; seven from the Low Carbon Energy (LCE) calls; six from the Energy-efficient Buildings (EeB) calls; four from the SME Instrument (second phase); and one each are funded under the ICT call 2015 and from the European Research Council (ERC) and Marie Skłodowska-Curie actions (MSCA).A total of 20 projects, almost half the total, have been funded under the contractual Public-Private Partnerships (cPPP).The majority of these ( 14) come under the Energy Efficiency calls, with the remaining six being funded under EeB calls.1.

Comparative Analysis
Figure 1 shows the number of projects supported by the different funding schemes while Figure 2 shows their total budget costs and the corresponding EU grant contribution.The majority of the projects are Innovation Actions (IA) and Research and Innovation Actions (RIA).However, when one looks at the allocation of funds, there is a large dominance of Innovation Actions accounting for some 82% of the EU grant contributions.This can be explained by the fact that the Innovation Actions include several projects funded under the Smart Cities and Communities calls, and these tend to have far larger budgets than other Horizon 2020 projects.Likewise, the almost negligible proportion of funds allocated to CSA, MSCA and ERC actions relates to the inherently smaller size of this kind of project.
The 42 projects have been funded across 28 topics within 23 calls for proposals from the years 2014, 2015 and 2016 (see Figure 3).Of the 42 projects, 15  The coordinators of the 42 projects are situated in only 13 countries, which are shown in Figure 4. Nine projects are coordinated from Spain and five each from the UK, Netherlands and Italy.France is next with four projects and the remainder are shared between Germany and Belgium (three each), Greece and Ireland (two each), and one each in Switzerland, Portugal, Sweden and Israel.Therefore, none of the projects is coordinated from former Eastern Bloc or so-called "EU13" [33] countries.Of the two coordinators who are based in non-EU countries, the Swiss one is an SME  The coordinators of the 42 projects are situated in only 13 countries, which are shown in Figure 4. Nine projects are coordinated from Spain and five each from the UK, Netherlands and Italy.France is next with four projects and the remainder are shared between Germany and Belgium (three each), Greece and Ireland (two each), and one each in Switzerland, Portugal, Sweden and Israel.Therefore, none of the projects is coordinated from former Eastern Bloc or so-called "EU13" [33] countries.Of the two coordinators who are based in non-EU countries, the Swiss one is an SME Instrument project (POWERCLOUD: a cloud energy management solution for office equipment and The coordinators of the 42 projects are situated in only 13 countries, which are shown in Figure 4. Nine projects are coordinated from Spain and five each from the UK, Netherlands and Italy.France is next with four projects and the remainder are shared between Germany and Belgium (three each), Greece and Ireland (two each), and one each in Switzerland, Portugal, Sweden and Israel.Therefore, none of the projects is coordinated from former Eastern Bloc or so-called "EU13" [33] countries.Of the two coordinators who are based in non-EU countries, the Swiss one is an SME Instrument project (POWERCLOUD: a cloud energy management solution for office equipment and smart devices) which is therefore acting alone rather than in a consortium, and the Israeli one is a European Research Council Proof of Concept grant (ERC PoC) for "Intelligent Control of Energy Storage for Smart Buildings and Grids".All of the remaining projects that include relatively large consortia are therefore coordinated from Western European countries.Figures 5-7 illustrate the extent to which tasks related to the various aspects of a smart building, as previously explained in Section 4 above, are spread across the identified projects.It is possible to conclude that 29 projects are exploring the user interface for control of a smart building, 25 projects deal with on-site storage, and 11 projects are investigating the links between smart buildings and electro-mobility and smart charging.If one considers the three main features of the proposed Smart Readiness Indicator that was referred to earlier, the results show that 39 out of the 42 projects are investigating the building's ability to interact with its users, 38 out of the 42 are investigating the building's ability to manage itself, and 32 out of the 42 projects are investigating the building's interaction with the wider energy environment.Thus, the three principal features are relatively equally represented across the range of projects.Nevertheless, within these totals it can also be said that the areas that are investigated by the fewest projects are electro-mobility and smart charging (12 projects), domestic appliances (11 projects) and self-learning/artificial intelligence (14 projects).
It should be stated that the budget and EU grant contribution figures quoted here represent the overall costs and contributions for the projects as a whole.In many cases, projects are working on a variety of tasks, not all of which are related directly to the advancement of smart buildings for sustainable energy.For example, actions funded under the Smart Cities and Communities (SCC) calls are very large and complex projects with a wide range of tasks that typically include developing digital networks and services for citizens, which may sometimes be related to smart use of energy in buildings, but not always.Likewise, projects working in the area of Demand Response may include tasks relevant to the energy consumption of buildings (which are included in this exercise) but they may also include tasks that focus on the wider energy grid outside of the building's plot boundary.Although such tasks do not feature in the mapping exercise, they nevertheless appear in the overall project budget costs.Figures 5-7 illustrate the extent to which tasks related to the various aspects of a smart building, as previously explained in Section 4 above, are spread across the identified projects.It is possible to conclude that 29 projects are exploring the user interface for control of a smart building, 25 projects deal with on-site storage, and 11 projects are investigating the links between smart buildings and electro-mobility and smart charging.If one considers the three main features of the proposed Smart Readiness Indicator that was referred to earlier, the results show that 39 out of the 42 projects are investigating the building's ability to interact with its users, 38 out of the 42 are investigating the building's ability to manage itself, and 32 out of the 42 projects are investigating the building's interaction with the wider energy environment.Thus, the three principal features are relatively equally represented across the range of projects.Nevertheless, within these totals it can also be said that the areas that are investigated by the fewest projects are electro-mobility and smart charging (12 projects), domestic appliances (11 projects) and self-learning/artificial intelligence (14 projects).
It should be stated that the budget and EU grant contribution figures quoted here represent the overall costs and contributions for the projects as a whole.In many cases, projects are working on a variety of tasks, not all of which are related directly to the advancement of smart buildings for sustainable energy.For example, actions funded under the Smart Cities and Communities (SCC) calls are very large and complex projects with a wide range of tasks that typically include developing digital networks and services for citizens, which may sometimes be related to smart use of energy in buildings, but not always.Likewise, projects working in the area of Demand Response may include tasks relevant to the energy consumption of buildings (which are included in this exercise) but they may also include tasks that focus on the wider energy grid outside of the building's plot boundary.Although such tasks do not feature in the mapping exercise, they nevertheless appear in the overall project budget costs.There are a number of reasons to challenge the results here presented.A manual database search is inherently open to human error.It is possible that requests for information that were sent to project coordinators and officers of the European Commission were not sufficiently clear and that they therefore could have been misinterpreted.It is also possible that the relevance of some project tasks was missed, or misunderstood, when analysing the project tasks.Some of the project documentation is very long and complex, and it is not always straightforward to extract simple yes/no answers.It is also possible that some projects were missed entirely in the search, for example if the search terms used did not appear in the data, including the abstracts, which are available on CORDIS.The results should therefore be treated with some caution.It is more likely that projects or individual tasks were missed, rather than being included in error.Because of this, one conclusion that can be stated with a fair degree of certainty is that, by early May 2017, at least 42 projects that included tasks related to research, innovation and market uptake of smart buildings had been funded under Horizon 2020.There are a number of reasons to challenge the results here presented.A manual database search is inherently open to human error.It is possible that requests for information that were sent to project coordinators and officers of the European Commission were not sufficiently clear and that they therefore could have been misinterpreted.It is also possible that the relevance of some project tasks was missed, or misunderstood, when analysing the project tasks.Some of the project documentation is very long and complex, and it is not always straightforward to extract simple yes/no answers.It is also possible that some projects were missed entirely in the search, for example if the search terms used did not appear in the data, including the abstracts, which are available on CORDIS.The results should therefore be treated with some caution.It is more likely that projects or individual tasks were missed, rather than being included in error.Because of this, one conclusion that can be stated with a fair degree of certainty is that, by early May 2017, at least 42 projects that included tasks related to research, innovation and market uptake of smart buildings had been funded under Horizon 2020.
New Horizon 2020 grant agreements continue to be signed on a regular basis.It is therefore quite likely that new projects that might be relevant to this exercise have been funded and have begun work during the period in between the search for projects and the subsequent analysis and writing of this paper.The results here presented should therefore be understood as a snapshot in time of the support given by Horizon 2020 to smart buildings research, innovation and market uptake from the beginning of the programme in 2014 until roughly its mid-way point in early May 2017.During the search for data it became apparent that a number of ongoing projects funded under the EU's 7th Framework Programme (FP7) [34], the predecessor of Horizon 2020, were also carrying out tasks related to smart buildings, but these were excluded from this exercise.The total level of EU financial support for smart buildings is therefore likely to be significantly greater than the figures that are presented here, especially when one considers the other possible sources of EU funding such as regional development [35] and Cohesion funds [36].
The analysis of data finds that no single Horizon 2020 project identified in this study includes tasks that comprehensively investigate all aspects of a smart building, according to the table with 16 features adopted by this study.However, a small number of projects do come close to comprehensive coverage.For example, projects GrowSmarter [37], AnyPLACE [38] and ELSA [39]  New Horizon 2020 grant agreements continue to be signed on a regular basis.It is therefore quite likely that new projects that might be relevant to this exercise have been funded and have begun work during the period in between the search for projects and the subsequent analysis and writing of this paper.The results here presented should therefore be understood as a snapshot in time of the support given by Horizon 2020 to smart buildings research, innovation and market uptake from the beginning of the programme in 2014 until roughly its mid-way point in early May 2017.During the search for data it became apparent that a number of ongoing projects funded under the EU's 7th Framework Programme (FP7) [34], the predecessor of Horizon 2020, were also carrying out tasks related to smart buildings, but these were excluded from this exercise.The total level of EU financial support for smart buildings is therefore likely to be significantly greater than the figures that are presented here, especially when one considers the other possible sources of EU funding such as regional development [35] and Cohesion funds [36].
The analysis of data finds that no single Horizon 2020 project identified in this study includes tasks that comprehensively investigate all aspects of a smart building, according to the table with 16 features adopted by this study.However, a small number of projects do come close to comprehensive coverage.For example, projects GrowSmarter [37], AnyPLACE [38] and ELSA [39] are found to include tasks covering 14 out of the 16 aspects of smart buildings.

Conclusions
This research work has found that the Horizon 2020 programme has supported 42 different actions that relate to smart buildings, between its beginning in 2014 and early May 2017.About half of these are Innovation Actions.The total budget costs for these 42 actions add up to 367.9 million Euros, of which the EU grant contribution is 304.1 million Euros.A number of these projects are only indirectly addressing smart buildings, for example through individual tasks or work packages within a large and complex project.In theory it would be possible to add up the budgets of all the relevant individual tasks and work packages to arrive at a precise figure for the funding of research, innovation and market uptake related to smart buildings in Horizon 2020; however, in practice such an exercise would be very complex and time consuming.
Despite the usefulness of central IT tools such as CORDIS, there is no automated way to identify projects working on smart buildings, a subject that cuts across Horizon 2020 topics and priority areas, and manual searches have been necessary.Similar efforts would be required to carry out an equivalent exercise in a separate cross-cutting discipline covered by Horizon 2020.The research has found that no single project is comprehensively investigating every aspect of smart buildings.However, three projects contain tasks that relate to 14 out of the 16 aspects of smart buildings that have been set out in this study.
This research could potentially lead to follow-up action, including policy discussions and further research.For example, it could potentially provide input to discussions at the level of the EU's Strategic Energy Technology (SET) Plan [7] which promotes alignment and cooperation in energy technology policies and funding priorities between EU countries, businesses and research bodies as well as the EU itself [7].More specifically, it could inform the SET-Plan Action 5 on new materials and technologies for buildings [40] and Action 3 on new technologies and services for consumers [40].The research could also potentially provide input into the Energy-efficient Buildings Public-Private Partnership [23], for example by providing data that could inform the next multi-annual roadmap and longer term strategy [26].In terms of the Horizon 2020 programme itself, the research might inform the remainder of the work programme and its possible successor EU programmes [41], as well as national funding programmes.Finally, individual research and innovation projects, as well as potential proposers of future projects, will be able to better understand the scope and scale of current efforts in this area.The research should help the project identify and make contact with others that are carrying out similar work, with the result that dissemination of results could be aligned for maximum impact.

4. 1 . 1 .
The Building's Ability to Interact with Its Users

Figure 1 .
Figure1.Number of projects per funding scheme.Source: author.For explanations of abbreviations see Table1.

Figure 2 .
Figure 2.Total budget costs and EU contribution per funding scheme.Source: author.For explanations of abbreviations see Table1.

Figure 3 .
Figure 3. Number of projects funded per Call.Source: author.

Figure 4 .
Figure 4. Number of projects per coordinator's country.Source: author.

Figure 4 .
Figure 4. Number of projects per coordinator's country.Source: author.

Figure 5 .
Figure 5. Number of Horizon 2020-funded projects that include tasks related to the ability of a building to interact with its users.Source: author.

Figure 6 .Figure 5 . 22 Figure 5 .
Figure 6.Number of Horizon 2020-funded projects that include tasks related to the ability of a building to manage itself.Source: author.

Figure 6 .Figure 6 .
Figure 6.Number of Horizon 2020-funded projects that include tasks related to the ability of a building to manage itself.Source: author.

Buildings 2017, 7 , 105 11 of 22 Figure 7 .
Figure 7. Number of Horizon 2020-funded projects that include tasks related to the ability of a building to interact with the wider energy environment.Source: author.

Figure 7 .
Figure 7. Number of Horizon 2020-funded projects that include tasks related to the ability of a building to interact with the wider energy environment.Source: author.

Table 1 .
Numbers of projects which contain tasks that directly relate to the energy smartness of buildings, and their total budgets and European Union (EU) contributions.Source: author.
Figure1.Number of projects per funding scheme.Source: author.For explanations of abbreviations see Table1.
have been funded under the Energy Efficiency calls; seven are from the Smart Cities and Communities calls; seven from the Low Carbon Energy (LCE) calls; six from the Energy-efficient Buildings (EeB) calls; four from the SME Instrument (second phase); and one each are funded under the ICT call 2015 and from the European Research Council (ERC) and Marie Skłodowska-Curie actions (MSCA).A total of 20 projects, almost half the total, have been funded under the contractual Public-Private Partnerships (cPPP).The majority of these (14) come under the Energy Efficiency calls, with the remaining six being funded under EeB calls.
Figure 2.Total budget costs and EU contribution per funding scheme.Source: author.For explanations of abbreviations see Table1. Figure 3. Number of projects funded per Call.Source: author.
Buildings 2017, 7, 105 9 of 22 smart devices) which is therefore acting alone rather than in a consortium, and the Israeli one is a European Research Council Proof of Concept grant (ERC PoC) for "Intelligent Control of Energy Storage for Smart Buildings and Grids".All of the remaining projects that include relatively large consortia are therefore coordinated from Western European countries.

Table A1 .
Matrix of Project Activities.

Table A2 .
List of Projects in Alphabetical Order.