Skip to main content

Analysing Transition planning and Systemic Energy Planning Tools for the implementation of the Energy Technology Information System

Final Report Summary - ATEST (Analysing Transition planning and Systemic Energy Planning Tools for the implementation of the Energy Technology Information System)

Executive summary:

To achieve the transition of Europe to a low carbon future, the Strategic Energy Technology Plan has been launched, focusing on strengthening and giving coherence to the overall effort in Europe in the direction of new energy technologies development and deployment. One of the main actions of SET-Plan is the establishment of an information system which will "map" technologies and provide updated information to support policy making (SETIS). In order to facilitate the development of the information system, ATEST brings together EU competence on the transition towards a sustainable and low carbon energy system through energy innovation, encompassing transition planning, energy modeling activities, and technology assessment. A workshop was held on 29th January 2010, in order to analyze the specifications (the needs that the models/tools will have to satisfy) within the scope of SETIS. An initial list of specifications was presented there, was discussed and further expanded. The audience of the workshop was SETIS (that is JRC and affiliations), Industrial groups and Governmental bodies from Member States. The outcome of the discussion with the stakeholders was a list of specifications for the models/tools that fall into four broad sections: Strategic Planning, Technology Deployment and Transition Planning, Innovation and R&D, International cooperation.

The final outcome of the ATEST project is a roadmap for tools necessary to plan and develop future energy systems and policies. In this framework a techno-economic and a socio-economic roadmap was developed based on drivers and suggesting actions in order to achieve the vision of advanced models and methodologies to support decision making related to the SetPlan implementation. An aggregated form of these roadmaps is presented in the report of WP6. All the reports of the project are available online at

Project Context and Objectives:
Over the last few years a growth in public concern about energy and environmental issues has appeared. Despite this wide concern Europe is not in the pathway to meet its policy goals regarding the environmental and energy issues. The challenges of climate change, security of energy supply and competitiveness are related and require a multidisciplinary and coordinated response. In order to meet the environmental and energy targets, the cost of clean energy technologies must decrease and the European industry must be the forerunner of the low carbon technology sector. In order to meet the greater ambition of reducing the greenhouse gas emissions by large amounts in 2050, new generations of technologies must be developed through breakthroughs in research. On one hand, the transition to a low carbon future, should prioritize the improvement of the European innovation system on energy technologies by creating at first the framework conditions and incentives for the development of new energy technologies. In order to achieve this, a formulation of a coherent strategy is necessary, coordinating the efforts of all Member States. On the other hand, effective strategic planning requires regular and reliable information and data. Essential components to provide this information are tools or methods that offer a better insight in technology transition analysis and planning, along with tools that focus on energy systems modeling.

The SET plan proposes a number of concrete actions to develop a more coherent energy research landscape in Europe. These include:
- Creating European Industrial Initiatives, focusing on technologies for which the barriers, scale of investment and risk can best be tackled collectively.
- Creating a European Energy Research Alliance, to enable greater co-operation across Europe of the research work going on in universities, research institutes and specialised centres.
- Planning the transition of European energy infrastructure networks and systems
- Creating a European Community Steering Group on Strategic Energy Technologies, which will allow Member States and the Commission to plan joint actions and coordinate policies and programmes.
- Establishment of an information system which will "map" technologies and provide updated information to support policy making (SETIS).

Effective strategic planning in the Steering Group requires regular and reliable information and data which will be provided by SETIS. Furthermore, SETIS is going to support the definition of energy technology objectives, as well as to build consensus around the SET-Plan programme.

The aim of the project was to create an initial framework as well as the roadmap for its future development, for providing answers regarding:
-the effect of policies to promote certain technologies on a country and EU level,
-the alternatives for meeting the energy and environmental targets that have been set and will be set in the future
-the effect of prices and taxes on the development of the energy system,
-the impact of Research Development and Demonstration on promoting and developing new energy systems,
-the issues related to the need for new infrastructure development for electricity, other energy carriers, carbon dioxide, and energy networks development in general.

The Objectives of the project are to:
-Define the questions that need to be answered regarding future low carbon energy systems.
-Review models/tools used in the European Countries, taking in mind what is used outside Europe, and what are the requirements of the SETPlan.
-Identify and recommend common tools and/or methods to be used in the MS and in the Energy Technology Information System (SETIS), and gain consensus on these models.
-Identify and recommend existing sets of data (on technologies, energy resources, statistics etc.), and provide a roadmap for the development of these data on a European and on a regional level.
-Identify the roadmap for the improvement and development of the tools and methods in order to cover the needs of the SETPlan implementation.

Objective 1: Review of models/tool

Objective 2: Identification and recommendation of common tools

Objective 3: Identification and recommendation of existing sets of data

Objective 4: Development of a roadmap for the improvement of the tools

Summarising the project will
-Evaluate the existing infrastructure and transition planning tools gaining a full insight in their structure and underlying theory.
-Identify the inefficiencies in existing modelling approaches
-Define the scientific and technical limits of transition tools
-Contribute to the advance of methods for the optimal routes for strengthening policies in infrastructure development.

Project Results:

The development of energy technologies plays a vital role in achieving the European targets of a 20% share of renewable energy sources in power generating capacity, 20% gain in energy efficiency and an overall reduction of 20% in energy use. The European Commission pursues to guide the transition of Europe to a low carbon future by launching the Strategic Energy Technology Plan (SET-Plan), which focuses on strengthening and giving coherence to the overall effort in Europe in the advancement of new energy technology development and deployment.

In the framework of the SET-Plan the European Commission has now initiated action on planning the transition of European energy infrastructure networks and systems with the launch of an FP7 Support Action named ATEST (Analysing Transition Planning and Systemic Energy Planning Tools for the implementation of the Energy Technology Information System). The aim of the ATEST project is to provide a ?toolbox? containing the methodologies, procedures and models required to support the decision making for the SET-Plan in the priority area of transition planning of the deployment of low carbon technologies and their supporting infrastructures.

Therefore, input was sought after regarding how energy policy decision making is done in practice. Based on this information a first course draft list of relevant questions and procedures was compiled. From discussions within the ATEST project team this draft list was refined and, as a result, it already hinted on a number of issues and overarching topics that need to be considered:

-trategic planning
-Deployment and planning the transition
-Innovation and R&D
-Reinforcing international cooperation

The final discussion and extension of the list of tool specifications was planned to take place through a workshop with a number of stakeholders involved in energy policy making and modeling on national and EU level. This workshop was also meant to provide in the need for transparency of the toolbox and to seek consensus between MS, Industrial Initiatives, the SET-Plan Steering Group, the modeling community and other relevant stakeholders. The final list of specifications resulting from these discussion and consultation efforts is presented here:

Strategic Planning

Key issues that were identified under strategic planning are the need for models to address both long- and short-term options to support the potential and opportunities of technologies in different stages, monitor whether industrial developments complement the strategic planning and provide means to perform bottleneck analyses. Technology performance with particular attention to potential for cost reductions will influence the deployment opportunities in the future and needs to be constantly monitored by means of suitable indicators. Also the interdependencies between energy technologies at various levels in the supply chain, the growth path of new technologies, and their overall impact on the energy system affect the deployment opportunities and need to be carefully considered. A better understanding is required regarding the effects of various policy instruments on the technology introduction.

Deployment and Planning of the Transition

After some strategic issues, deployment and planning the transition is looking into more applied issues that have to be taken into account during the introduction of new energy technologies. Spatial planning is concerned with the best suitable deployment locations and which infrastructure issues (e.g. connections) are required to harness the full benefits. The choice of demonstration projects should already incorporate the necessary framework conditions (e.g. logistics) to anticipate a further ramp-up of technology deployment by using the demonstration projects as seeds. Barriers and time lags related to investment decisions need to be understood. Public acceptance of technologies is an influencing factor with society that could possibly hamper technology deployment and it needs to be investigated how risks are perceived and if better ways of implementation are feasible.

Innovation and R&D

Innovation and R&D are topics that are strongly related to each other. The effects of R&D spending, respectively the effect of public and private R&D need further study for the future. Setting targets and monitoring progress for R&D are necessary in order to utilize available funds as effectively as possible. From an innovation perspective, the identification of European champions in energy technologies can help to re-focus resources and help to position EU in relation to the rest of the world.

International Cooperation

Improved international cooperation on energy technologies can help to stimulate knowledge development and accelerate technology development by sharing costs and benefits. So far, the effects of international cooperation on technology R&D are not properly analyzed and thus require investigation. Therefore evaluating which areas are potentially most effective for international cooperation and subsequently targeting these areas for increasing knowledge exchange may help to increase overall strength of the European Union's position.

General Concerns

The discussions, particularly those during the workshop, showed that the effectiveness of R&D funding and the role of public and private R&D are a major concern. The role and the effectiveness of policy instruments in bringing down costs and stimulating deployment as well as the understanding of the impact of a specific individual technology via policy indicators drew much attention during the workshop. This also applies to the interaction between international cooperation and competition. The technology learning curve was frequently mentioned as an instrument for the analysis of the expected cost reduction potential.

An inventory of existing models and tools that cover transition planning and systemic energy modelling was performed during Work-package 2 of the project. The main instrument to perform this task was through an open consultation, using a classification form (or questionnaire) and detailed guidelines for EU and non-EU modeling teams.

The different types of models and tools can be briefly described as follows:

-Disaggregated models are very detailed models that address specific issues such as plant design, resource potential assessment, infrastructure expansion or reinforcement, etc. Results from these models can generally be used as inputs to other more systemic models and can provide a sound reference for constraints such as capital cost, efficiency, resources, barriers, and so on.
-Sector level models are used to analyze parts of the energy system at different levels of detail: for instance, models for grid operation simulation or for the electricity system by itself, or models for single markets (coal, gas, oil, etc.) or for single sectors (transport, residential, etc.).
-Energy system models are intended to address and analyze the evolution of the energy system by combining its different parts (multiple sectors and fuels) with a focus on competition and complementarities between energy technologies.
-Macro-economic models include both the energy system and the rest of the economy possibly with feedback effects between them. Typically, the energy system is described as one component of the economic system (with obvious implications concerning the level of detail of the former).
-Energy behavior tools are designed mainly to make people aware of their energy consumption decisions (with a focus on the demand side of the economy). This category also includes semi-quantitative tools dealing with social acceptance issues.
-Socio-technical scenarios (STSc) address the way transition paths may unfold in a process of interaction between a range of actors and the rules they act upon (technical, regulatory, forms of provision, cost models, infrastructure requirements, etc.). The purpose of STSc is to illustrate how various transition routes may be set in motion through a variety of multi-level linkage patterns.
-Horizon Scanning methodologies point to the systematic examination of potential threats, opportunities and likely future developments, including (but not restricted to) those at the margin of current thinking and planning. The aim is to identify the potential impacts of wild cards (WI) and weak signals (WE) on Europe and the world: WI are situations/events with perceived low probability of occurrence but potentially high impact if they were to occur; WE are unclear observables warning people about the probability of future events (including WI).

The models analysed in this project have been grouped according to their analytical approach as well. The different analytical approaches that have been used to characterize models can be briefly described as follows:
- Bottom-up: models traditionally technology-oriented, treating energy demand as either given, for example expressed as useful energy demand, or as a function of energy prices and national income. Technologies are typically described as a set of linear activity models based on engineering data of life cycle costs and thermodynamic efficiencies.
- Top-down: models with primary focus on market and economy-wide feedbacks and interactions, often sacrificing the technological richness of the bottom-up approach; typically they represent technology by using relatively aggregated production functions for each sector of the economy.
- Hybrid: mainly top-down models that include some technological or environmental explicitness.
- Hybrid - Integrated Assessment: mainly bottom-up models with high technologic details and combining economic, technological and environmental details, mostly used to evaluate the impact of climate change policies portrayed by a damage function", but also models with an endogenous climate module representing the impacts of climate change in physical terms.
- Semi-Quantitative: analytical approach that considers not only quantitative aspects, but also qualitative ones, such as social acceptance issues related to energy projects.
- Qualitative: analytical approach for uncountable aspects of the energy system transition, such as social, energy system governance or policy-planning issues.

It should be noticed that there is a prevalence of technology-rich models. Disaggregated models (green area), Sector level models (pink area) and Energy system models (orange area) usually are rich in terms of technology details and their geographical coverage varies from local to global depending on the type. The blue area identifies the macro-economic type of models that are characterized mainly by a poorer technology detail and by a multi-country to global geographical coverage.

Strategic Planning

A long list of models addresses most of the specifications under Strategic Planning as their primary focus of analysis. This is true both for General specifications concerning the resilience of the energy system against different sources of shocks and for specifications related to ?Technology performance and development potential. Similarly, all the specifications under the heading Policy indicators (except for LCA analyses) are well covered by different types of identified models.

The most critical deficiency lies on the scarcity of models designed to investigate the possible ?Bottlenecks to technology deployment?. In fact, only one model (focused on hydrogen) aims at directly analyzing this type of problems. Although there exist models that take somehow into account constraints to technology penetration and diffusion (e.g. Markal-type bottom-up models), a more accurate coverage of this specification would require at least a deep analysis of the production chain for the technology under consideration.

Technology Deployment and Transition planning

Most of the specifications are appropriately covered by different types of models. For instance, Capacity expansion (infrastructure) and Grid-connection capacity represent the primary focus of some disaggregated models focused on infrastructure expansion and used by national TSOs and DSOs for electricity grid planning. Many models with different analytical approaches are useful to investigate specifications such as SET-Plan sectoral targets, whereas Links between the energy system and the economy are better analyzed by macro-economic models. The latter, together with energy behavior and other qualitative tools, are also helpful to cope with Market design and organisational changes and a few Acceptance/perception of a technology specifications.

Innovation and R&D

A good number of models covers most of the specifications related to ?Innovation?. On the other hand, R&D issues represent one of the most challenging tasks for modellers and for the scientific community. Qualitative methodologies represent the only appropriate tool to analyze the Risks involved in research activities from a long-term perspective. Moreover, there is a lack of useful models specifically designed to deal with technology specific R&D targets and monitoring of funding mechanisms or able to quantify the necessary amount of R&D spending needed to become or to stay competitive with non-EU countries. This is mainly due to the fact that R&D is an intrinsically uncertain activity and this makes it difficult to provide forecasts for specific technology needs and existing models are mostly based on historical data and trends on R&D funding.

In general, addressing R&D specifications from a modeling viewpoint is challenging because of several issues:
- data availability (e.g. need of global data, private data lacking, feedback between private R&D investments and sales, lack of global market model for technology production overtime);
- methodology (e.g. the learning curves is one possible approach, but it is necessary to account not only for the effect of deployment and R&D, but also for bottleneck issues);
- modeling coupling as feedback loops and spill-over.

International Cooperation

The detailed list of specifications and the analysis of a large inventory of models, described above, showed that in order to answer any kind of policy question about the future of the energy system, one needs to use a combination of several models/tools, and this combination can vary depending on the type of question or the characteristics of the Member State (i.e. existing energy system and policies, climate conditions, land use, etc.). This conclusion led to the creation of a methodology that can be followed in order to find the best available combination of models that can be used in order to answer a specific policy question. One of the main decisions that affect the combination formulation is the maximum number of models that can be used in a combination, which was limited to six in our analysis, in order to limit the complexity and difficulty of model linking. The methodology was applied to a list of relevant policy questions that were formulated by the project consortium, in order to demonstrate its functionality. In order to give the flexibility of adding more models into the analysis a software tool was developed, although it was not foreseen initially in the project.

The scope of the tools selection methodology is to come up with a combination of models and tools, giving guidelines on how to choose the best available set, depending on the policy question that needs to be answered. In this sense in order to answer a policy question one needs to combine specifications, find models that can answer to these specifications and combine them appropriately in order to give a final answer to the policy question. Two definitions are required in the application of the methodology:

-The usefulness of a model regarding a specification, which expresses how well the model can answer to this specification.
-The importance of a specification relative to a specific policy question, which measures the relevance of this specification in answering the policy question.

The methodology proposed follows these steps:
-Setting the scales for the quantification of the parameters
-Ranking models according to their usefulness in answering given specifications
-Ranking specifications according to their ability to answer policy questions
-Identifying combinations of models
-Evaluation of the ability of a combination of models to answer a policy question

Step 1: Setting the scales for the quantification of the parameters

Regarding the usefulness of the models, the following linguistic values scale was proposed to answer the question:

What is the usefulness of the model in addressing a given specification?

Individuals are likely to have different perceptions on what they mean by defining the usefulness of a model with none, poor, medium, good and very good. In order to account for this fact, for each of these linguistic weightings a lower (a), an upper (c) and a median (b) value is assigned.

Regarding the importance weighting of a specification relative to a specific policy question, the following scale was proposed to answer the question:

What is the importance of a specification in answering a given policy question?:
In order to quantify each one of the words in the scale (2), and to translate the linguistic information into a range that can be used in next step rating, a fuzzy set is associated with each one of the linguistic values. As for the previous parameter, three numbers (a,b,c) in the interval [0,10] are needed, in order to define it. Since the importance weighting of a specification is a weighting factor, it is defined as a positive value. The definitions of the range for the usefulness of the models and the importance weighting of the specifications have been discussed among the project partners and a consensus has been reached among them to use the sets describe bellow. This was done in order to have a common model ranking under a common framework among the project partners.

Step 2: Ranking models according to their usefulness in answering given specifications

The aim of this step is to rank each model/tool in respect to the quality of answers it can give to each one of the specifications. This step is independent of the policy question that needs to be answered. The Models Characterisation Report (WP2) presents for each of the models/tools, its ability to answer a specification or not, along with its primary focus specification. The analysis was done using the set of specifications that were derived from the public consultation and are presented in the Specifications Report (WP1).

For each of the models/tool, the project experts assigned values for the usefulness of the model in answering each one of the specifications in the new list. The ranking was based on the literature review for the model use (proven capabilities) and/or references for model description, including the knowledge of the models by the project team. During this process each model was ranked independently by two teams of project experts. The two rankings were then compared, discussed and a common ranking was reached. These model rankings were then sent back to the modeling teams and their feedback was requested.

Step 3: Ranking specifications according to their ability to answer policy questions

This step depends on the specific policy question that needs to be answered. As was mentioned before, each policy question can be related to a number of specifications. Since the goal of this approach is to create a methodology to select the most suitable toolbox, depending on the policy question, each specification will have a different weight in providing an answer to this particular question. That is the importance of each specification in answering the policy question is different, and some specification might not even be relevant. So each specification will receive a characterization from the scale (2), N, VL, L, M, H, which will be different for different policy questions.

The project team has set up a number of typical policy questions that were used for the testing of the methodology. The list of these questions is:
-How to achieve a low cost and low emissions energy mix?
-How to achieve an energy mix that maximizes employment opportunities?
-How to achieve an energy mix that has the maximum societal acceptance?
-Which are the most competitive low carbon technologies in the medium and long term?
-Where should new energy installations be best located?
-In which R&D areas should a country invest?
-How should a country develop energy interconnections with other European and non European countries?
-How to improve energy efficiency?

It is evident that these are general questions that require the combination of a number of specifications in order to be answered. So, for each one of these questions the specifications required must be identified and their importance in providing an answer must be ranked. In order to check the methodology the questions above were analysed and the importance of each specification in answering each of these questions was characterised.

Step 4: Identifying combinations of models

Once step 3 is completed for a given policy question, the next step is to set up combinations of models/tools that can provide answers to the required specifications, since it is unlikely that one model/tool can provide all the answers. The process is to capture the sufficient and relevant level of technology detail, sector and geographical coverage for each policy question and use this to select the model combinations required. The set of model combinations will be unique for each policy question. A detail presentation of the combination creation methodology can be found in the Models/Tools Selection Methodology report.

Step 5 Evaluation of the ability of a combination of models to answer a policy question

This is the final step of the methodology in order to evaluate the ability of a combination of models/tools to answer to a policy question. In this step a Decision Matrix is created with the specifications on one dimension and the model/tools combinations on the other. The weighted sum approach is applied to this decision matrix in order to find the preferred alternative among the combinations, for the specific policy question.

Following the methodology described above and the analysis and ranking, the results of the methodology for the one example policy question is given here and more examples are presented in the Models/Tools Selection Methodology report.

The availability and quality of the data used in the models are critical for the model outputs. Therefore an overview and inventory of data sources used in the different types of model families and the actual data taken from these sources were analysed in the next step of the project. All data which are used were evaluated according to their quality and availability and the weak points of the existing data were described. Finally, the additional data requirements to improve the models themselves and therewith the quality of their outcomes, are also described and ways to generate the missing data were discussed. To organize the evaluation of the data sources and the analysis of the additional requirements, the work is structured according to the different model families. This clustering is related to the model types described above.

For an evaluation of the data sources and data used by the different models, a very deep knowledge of the models or tools is needed. For this reason, only specialists deeply involved in the models generation could be asked to perform the data evaluation for their own model. For practical reasons (ATEST covers 85 models) not all models can be assessed individually. However this work attempts to cover the data and information and also the additional data needs of the model or tool types described in the ?Models Characterisation Report? by relating them to model families, instead of single models. The model families used are: Energy System Models, Macroeconomic Models, Sector Level Models, Disaggregated Models, Energy Behavior Tools, Socio-Technical Scenarios and Horizon Scanning Methodologies. The models within the same family share a similar structure and use comparable data. Therefore the additional needs are connected to the same family type. Each model family is represented by one key model example which is analysed in detail. Other models within the same model family were also considered if data and expert knowledge were available. Data issues found for the analysed models were assumed to hold for all models within that family. In addition to qualitative models, also the information used by different types of tools is highlighted. Typically the above mentioned Energy Behavior, Socio-Technical and Horizon Scanning tools and methodologies use also qualitative information, which might be gathered by stakeholder interviews or different surveys, for example.

The work of collecting and evaluating the existing data sources and data itself as well as the analysis of the additional requirements is split among the ATEST project partners to gain expertise for different model types. In addition to the work of the ATEST project partners, an Open Call was launched in order to include the input of other modeling teams (called external modeling teams in this report).

To structure these three steps, i.e. data source collection, existing data evaluation and data needs analysis, a data collection list was developed. The results obtained from this list are presented in detail in the Existing Data and Data Requirement report.

The data sources and information were described using different parameter categories (with either predefined answers to choose from or free text answer):
-Data Source:
-Name of the data source [free text]
-Classification of the data source [predefined]
-Category of data, split in main and additional (if needed) category of data [predefined]
-Description of the data taken from the source (like primary energy consumption by fuel/sector/country, capacities,?) [free text]
-Sector coverage of the data
-Sector which is covered [predefined]
-Additional comments on the sector coverage if needed [free text]
-Geographical resolution of data [predefined]

The objective of this part of the data list is to create an inventory of the various sources and describe the data taken from these sources. Therefore it is important to clearly point out which specific data were taken from which source.

In the second step, the data were evaluated based on the description of the sources. The data assessment is performed using different indicators which measure the quality and the availability of the data, and identify the weak points of the existing data. The evaluation should be the basis to point out the weak points of the current data. Based on these shortcomings, in the next step additional data requirements are described to close these data gaps.

The categories and sub-categories to assess the existing data are as follows:
-Quality of data:
-Time resolution of the provided data [predefined]
-Time lag (between date of publication and reference date) [predefined]
-Frequency of data publishing [predefined]
-Level of completeness and consistency [predefined]
-Level of detail [predefined]
-File format [predefined]
-Data documentation [predefined]
-Status of availability [predefined]
-Price [free text]
-Description of weak points of existing data [free text]

The evaluation is performed by model experts. Concerning the evaluation of the data according to their completeness/consistency and the level of detail, additional guidelines were worked out. These guidelines were provided to the experts from modeling teams among the ATEST project partners and to the external modeling teams within the framework of the Open Call.

The categories and sub-categories of data needed are as follows:
-Lack of data:
-Sector covered by required data [predefined]
-Additional comments on the sectoral coverage (if needed) [free text]
-Geographical resolution of the required data [predefined]
-Description of the lack of data/ additional data needs [free text]

In total 226 sources and additional data needs were analysed in this study. The analysed data give an overview about the currently used data in the different models and also show the weak points of these data sets as well as the additional data requirements.

The TIMES PanEU model covers all Member States of the EU-27 plus Norway, Iceland and Switzerland and is technology rich. Therefore many data sources are needed to model the different sectors in each country. EnergyPlan is developed by the sustainable energy planning research group at the University of Aalborg (Denmark). It is a free program for energy system analysis with different input and output tables and the main target is to assist the design of national energy planning strategies.

But also literature and information from associations play a key role as a source. The data from associations are mainly used in the industrial sector which is also modeled in detail in the model example TIMES PanEU. Looking at Macroeconomic Models in this study, the data sources are dominated clearly by official public statistics.

Energy Behavior Tools use in contrast to the other models and tools behavioral information, qualitative information in interests, values and expectations as well as qualitative information on context. According to their model character, the technology rich Energy System Models mainly use energy and technology input data while the Macroeconomic Models use mainly macroeconomic data. The other important category at this model type is other model input which includes general information about population or exchange rates.

This report analyses also the quality of the data sources and attempts to highlight the weak points by model family. One indicator to measure the quality is the time resolution of the provided data. The dominating resolution in most of the models is the annual level (77 % of all sources). However, the time resolution of the information used in Energy Behavior Tools depends on the issue of focus (all sources used in Energy Behavior Tools). This indicates the particular nature of this model family. A more detailed time resolution is relevant at Disaggregated Models and at the Energy System Model EnergyPLAN.

Concerning the time lag between the date of publishing and the reference date, most information used in the models have a time lag between one and two years (48 % off all sources) followed by less than a year (23 %). The Energy System Models analysed here also use future-oriented data, about 20 % of the data sets used in Energy System Models are of this kind (12 % of all sources used in all models). They are needed for the demand calibration or concern the expectations about technological developments or renewable potentials.

The frequency of data publishing evaluates also the data quality. The data analysed are dominated by yearly updated data (52 %). Differing from this frequency, the Sector Level models of this study use mainly data which are continuously updated which could make it more difficult to stay up to date with the model. Disaggregated Models and Energy Behavior Tools also use data which are not periodically updated (31 % of the data used in Disaggregated Models; 33 % in Energy Behavior Tools). Also this frequency makes it harder to keep the model calibrated to the lATEST publications. The majority of sources used in Energy Behavior Tools are not published (67 % of the sources used in Energy Behavior Tools). This refers to the way of data collection via interviews or workshops.

The analysis of the data weak points was performed based on the information collected. General weak points are that some sources which are needed for every region of the model are available only for one country and in the specific language of this country. As a result, not all model regions are calibrated with the same quality which could have an impact for example on the quality of the model results concerning the burden sharing of emission reduction targets. This affects mainly Energy System Models. Other data sets exist just in aggregated form and the modeling team has to split the information into the regions covered by the model, based on assumptions. Concerning regional aspects, another issue is that national statistics often do not fit to international statistics and in general data from different sources rarely fit to each other. The use of different balancing rules seems to be one reason for that. Moreover, some sources are not free of charge.

The analysis addresses also the issue of the additional data requirements, together with the weak points identified by model family. The analysis of Energy System Models shows that this model family needs many different sources because of the detailed modeling of different regions and sectors. A specific challenge is that all data are needed for all regions covered by the model. This strong point of multi-regional models often becomes a weak point. Furthermore, the use of different balancing methods between the different regions might become challenging. The additional data requirements of Energy System Models point out firstly the need for technological information about the existing stocks, such as the efficiency or lifetime, and additional information on technology cost, especially for modeling the demand sectors, like the industrial sector. Additional needs are addressed in the CHP modelling with concern to data on electricity, heat and steam production split by technology, energy carrier, autoproducer or public CHP, and industrial sub-sector.

For a model extension or improvement, there are data needs concerning energy balances of useful energy, like the demand for space heat, process heat or cooling. A key requirement made by the modeling teams of this family relates to additional data with a higher time resolution to enable modelers to include more time slices in the model; and with a larger geographical coverage to can include new regions, e.g. Turkey and Croatia. To that, the implementation of a higher time resolution needs special load curve data of different energy commodities (like electricity and heat) and different sectors.

The analysis of Macroeconomic models (like NEWAGE, GEM-E3 and GTAP-E) shows that these models are based in general on highly aggregated data. This is an essential property, but often misses details on technology, preferences or environmental impacts, which play a crucial role for the outcome of specific macroeconomic topics such as the macroeconomic impacts of environmental policy instruments, the role of capital and energy as main input factors in the production and the consumption of energy services.

The analysis performed within this Work Package joins the findings of the previous Work Packages that each Policy Question should be answered with different types of models and/or tools. From data viewpoint, each model type employs different ways of generating the required information. To improve the existing model platform, additional data are required and the availability of high quality data is a key issue for the future development of these models.

Particularly important data attributes for model improvements and enlargements are the availability of more detailed data, e.g. higher time and geographical resolutions, the data consistency and comparability of databases among countries, regions and sectors. On the other hand, depth of information and context-specific understanding may be even more important when changes in behavior are strived for.

As was mentioned above, the linking of models is necessary in order to answer a Policy question. In order to study how can this be done in practise some models/tools were used in a demonstration. The models/tools which were selected to be used are TIMES, COMPETES, Climate Bonus, RESolve-E, IER-Transmission and MECHanisms. The main reasons for selecting these five specific models and tools for the demonstration exercise were the fact that they appeared in the top combinations of WP3 and at the same time they were available to the project partners. In addition, the combination of these models and tools were expected to highlight:
-The possibility of linking models that use different time resolutions,
-The problems/advantages of linking overall energy system models with specialized models of the electricity system,
-The possibility of linking techno-economic with behavioral models.

Some of these models are run on an EU level, while others apply on a national scale. The models use different time scales for the analysis, i.e. hour, year or multiyear. As a result, all the data have to be formed so as to fit adequately to each model. Moreover the issue of how to synchronize independent model runs must be addressed, especially when the models have to exchange information between them.

Most of the models have a significant amount of common input data. So, we have assumed the existence of a common database (CDB) from which the models extract their inputs. The advantage of having a common database covers the need for maximum possible data consistency. This database should include all the necessary data at the highest available geographical granularity and time scale, in order to allow the extraction of data at the level desired by each model. The rest of the data input for each model is assumed to be stored in independent, dedicated databases.

The remaining models focus on specific areas: three of them (RESOLVE-E, COMPETES, IER-Transmission) are models strongly related to aspects of the electricity system, while the other two are more behavior-oriented (Climate Bonus relates to individual- and aggregated-level consumer behavior and MECHANISMS for interpretation of behavioral issues in specific project contexts). Therefore, we firstly address linkages between RESOLVE-E and COMPETES, and in parallel the model IER-Transmission; the outputs of these three models refine the power sector attributes that are used as an input to the TIMES model. More specifically, the linking scheme could follow the steps below:

Step 1:

CLIMATE-BONUS provides enhanced CO2 emission data for the consumption sectors. The model can calculate CO2 emissions per energy use, based on detailed data of technologies and behavior. It can therefore provide more accurate coefficients that can be used in a systemic model like TIMES, enhancing in this way the representation of GHG emissions that are not dependent only on technologies but also on behavioral aspects. Furthermore, the selection of a representative sample of the Climate Bonus database can be used as a reliable basis for the calculation of consumer elasticities related to the price or quantity of CO2. Using the results of the application of MECHanisms, in a large number of cases/projects, important information may be obtained regarding typical social behaviors in various energy efficiency programs/projects. Such kind of information provides an opportunity of quantifying different modes of social involvement, especially when compared to BAU conditions, and critical insight can be gained, especially on the modeling of social behavior. This is the kind of input that MECHanisms will provide to TIMES. Nevertheless, testing these assumptions on real case studies is considered crucial for determining the connections between MECHANISMS and the other models. The electricity demand from TIMES is used as an input to IER-Transimission. IER-Transmission also uses other input from its own independent database and the common database, while the output of the model (thermal power investment plan, and power transmission investment plan) feeds TIMES, in the form of forced investment scenarios.

Step 2:

The electricity demand provided by TIMES is used as an input to the iteration between RESolve-E and COMPETES (which focus on the power sector and renewables in particular). The thermal power system and transmission is optimized by IER-Transmission.

According to the depicted scheme, RESolve-E and COMPETES will be used to fine tune the operation of the power sector. Both these models have been developed at ECN, and they are coupled in the framework of WINDSPEED project [5]. Based on this experience, it was attempted to extend further this coupling framework inside the linking scheme presented here. TIMES which as an energy system model represents endogenously the competition between the different energy carriers (natural, gas, oil electricity, etc) and therefore provides the final electricity demand to RESolve-E & COMPETES. For RESolve-E, the annual final electricity demand is a direct input, but in COMPETES further analysis of the annual final electricity demand is needed. It is necessary to transform the annual final electricity demand to an hourly pattern, since in COMPETES an hourly simulation is conducted.

RESolve-E requires as driver-input the wholesale electricity price, and during the first run of RESolve-E, the marginal price of electricity derived with TIMES can be used. The other basic inputs required for the first run of RESolve-E, coming either from the independent (IDB) or the common database (CDB), are support schemes for RES capacity development (IDB), fuel prices (CDB) and technical and economic characteristics of RES technologies. It is important to note at this point that RESOLVE-E has to be properly calibrated with the relevant historical data of RES capacity expansion before running the model.

Once the first run of RESolve-E is finished, COMPETES can be run using the RES capacity development and RES electricity production taken from RESolve-E, the final electricity demand taken from TIMES, the thermal generation investment plant and the transmission investment plan taken from TIMES, shaped by the output of IER-Transmission. IER Transmission has been chosen for this role because it is formulated to optimize both generation and transmission capacity expansion simultaneously. Furthermore as one is approaching 2050, the role of thermal generation is highly transformed to complement RES electricity production. This transformation will be highly proportional to the lowering of the average cost of RES production, the appearance of commercial storage technologies and the high levels of CO2 emission costs.

Step 3:

The remaining input data come from the common database and an independent specialized database. A series of the most important information that will shape the power subsector of TIMES are included in this table. These include the RES capacity development and RES electricity production, RES electricity curtailment, power trade flows and congestion prices (the marginal price of the transmission capacity). With these updated inputs TIMES is run again, providing an update of the development of the overall energy system and of the electricity demand.

A convergence scheme is required for this process, which is necessary since the iteration between the model runs will result to a shift of the optimal points in each model. In order to deal with this we need to define a kind of ?confidence intervals? between some key exchange variables. One such key variable is the final electricity demand. TIMES provides the final electricity demand to RESOLVE-E, COMPETES and IER-Transmission, and RESOLVE-E returns cost development indications about renewable technologies while COMPETES provides the wholesale electricity price to RESOLVE-E and IER-Transmission provides the thermal power and transmission grid development. It is highly probable that for the interaction scheme of Figure 10, more sophisticated convergence rules are required, but these can be derived only during the process of actually linking the models. The optimality of the solution in such an it eration is not guaranteed, and it should be closely monitored and empirically verified. The convergence of the results of the different models for the same variables (e.g. power flow calculated from IER-Transmission and COMPETES converging to the same value) is an additional sign that the linking process approaches a common solution.

The final outcome of the ATEST project (WP6) was to create framework and a roadmap for tools, models and methodologies necessary to plan and develop future energy systems and policies. The transition to low carbon energy systems requires a paradigm change and thereby analysing the transition requires multilevel and multidisciplinary perspectives to formulate and analyse effective policies and strategies. The analysis made within ATEST WP2-WP4 clearly revealed that the existing models and tools and their databases would not cover the multilevel perspective (MLP) of the transition process. In the MLP approach, the assessments consider three levels: energy infrastructures (i.e. energy systems and technologies), behavior (i.e. consumer's and investor's chooses), and institutional factors (i.e. policy, regulation, and markets). These three levels form a regime level of the specified system. On the other hand, the depiction of the regime level of the system is put in the governance context covering the phases of Goal formulation, System analysis and Strategy implementation. The formulation the ATEST framework was based on the above theories. The framework was presented as a matrix to identify the gaps in existing tools, models, and methodologies. A color coding is added so that it is easy to recognize the gaps in models and methodologies development. Green color means that the area is well covered with several tools or models, yellow refers to moderate coverage with some deficiencies, orange to coverage with serious deficiencies, and red refers to areas that are not covered at all.

The major findings identified by the ATEST framework including the above country/EU studies may be summarized as follows:

-Existing policy making among EU MS usually includes some background studies assessed with energy system and/or macroeconomic models. Instead, behavior and institutional aspects were barely considered in the policy making.
-Analysis of the transition of the energy infrastructures is well covered with existing tools and methodologies. The greATEST gas was identified in tools and methodologies focused on analysing the effectiveness of RD&D policies, consumer and/or investor behavior and institutional factors. On the other hand, energy system modeling barely takes these issues into account.
-Transition to low carbon societies call for new traditions in policy making process and new tools and methodologies to foster effective SET-Plan implementation. Increasing complexity of tomorrow's energy system requires more complex models and/or integration of several models/tools and thereby modifications to existing models.

Based on the gaps identified in the ATEST framework roadmapping exercise for tools and methodologies development was conducted among the ATEST partners. In the ATEST project, and most often in also other roadmap processes, the technique is applied in through a combination of group- and desk-work. The process requires inputs from people with deep knowledge about the focus area. The core idea is to formulate future actions (step by step path) in order to attain the wanted future (vision). The vision statement is a crucial part of a roadmap as it defines the goal for the activities described in the process. For the ATEST roadmap the vision was derived from the project plan and it was edited during the working process. The final formulation of the vision statement is as follows:

Advanced models and methodologies to support decision making related to SET-Plan implementation

The ATEST Roadmap work was carried out in five different phases:
-Preparation of a roadmap template.
-Generation of roadmap content in internal workshops by each ATEST partners.
-Prioritization of roadmap content in a video conference by all ATEST partners.
-Composition of the roadmaps by VTT project group.
-Commenting and edition of the roadmaps in a workshop organized in ATEST final conference.

The roadmap template used in the ATEST project has four different layers. The top layer describes the drivers and challenges at the global and EU level. The next layer emphasizes the technological development, infrastructures and the behavior of consumers and investors affecting the models and methodologies development. The third layer shows the expectations of the model development. The last layer of the roadmap addresses the actions needed in order to achieve the vision taking into account the development that were identified in the upper layers of the roadmap. After the steps 1-3, it was evident that there were so much content generated in the process that it was not possible to include everything in a single roadmap. Therefore, the content was divided into two separate roadmaps that address the development from a) technoeconomic and b) socioeconomic viewpoints. The final step of the roadmap process was to finalise the visual roadmaps together with the participants of the final ATEST workshop in Brussels in 26 March 2012.

According to the techno-economic roadmap, a special emphasis should be put on the following issues:
-More transparent modelling in supporting of the EU RD&D policy initiatives. This was considered an overall action, which should be taken into account in realizing the following actions as well.
-Standardized scenario-based benchmarking of models, common standards of reporting (including uncertainties) should be realized as an early action.
-Setting-up of common and transparent databases between models and integration of real time data (i.e. statistics) would be recommended as a short term action as well.
-Research program on integrated assessment of models, tools, methodologies, and databases, which should include:
-model and database coupling techniques
-better tools for assessment of environmental impacts
-better approaches to link energy system and economic modeling
-geographic and land use models
-modeling of infrastructures

In order to gain development in decision making in implementing the EU Set-Plan by modeling and tools, according to the socio-economic roadmap process, actions needed touch behavioral and institutional, namely R&D policy, factors in the regime level. Actions recommended from the point of view of socio-economic roadmap are as follows:
-monitor systematically the effectiveness of RD&D policies (include both ex-ante and ex-post analysis)
-highlight the big issues, which have the direct impact on SetPlan im-plementation
-create tools to analyse the R&D spending, consumer behavior and market agents
-implement the tools and methodologies for stakeholder involvement

In reality, there exist an unlimited number of relevant policy questions, among them also questions that focus on the ways in which keyactors' behavior, attitudes and acceptance can be influenced (limited attention has so far been paid to this type of contribution). It is thus good to keep in mind that new types of policy questions linked to other types of specifications - may emerge. This emphasizes the need for further work in figuring out reasonable and useful ways of combining models and tools. This was a clear output of both socio-economic and techno-economic roadmaps, which included the same action, i.e.:
-integrate the outputs of socioeconomic actions in decision-making via techno-economic models, via feedback between techno-economic and socioeconomic models, or directly in the policymaking process via some other way.

Potential Impact:

The acceleration of the development and market take up of low carbon energy technologies is one of the main strategic objectives of the European Energy Policy. In order to meet this objective existing structural weaknesses must be overcome, in order to catalyze energy technology innovation. The challenge is to do this in a way that maximizes the potential competitiveness gains for Europe, and limits the potential costs. Among the necessary preconditions is the availability of tools that will be used in order to provide better insight on the effects of different policy options. The outcome of the tools will be used for optimizing the allocation of energy technology innovation process, formulating a coherent, stable framework for energy technology innovation by putting the grounds for a sound and promising EU energy strategy in technology as a whole. A number of tools exist today on systemic energy modeling and energy transition planning that have been developed independently from each other, which sometimes resulted in the lack of transparency.

The project is expected to support the decision making for the development of low carbon energy systems, through the review, analysis and evaluation of the existing tools and at the same time suggestions for future developments of specialized tools (models/methodologies) that will improve the analysis capability on a Member State level and on the EU level. The evaluation criteria used in this process, were directly associated with the objectives of the SETPlan and the Energy Technology Information System in particular.

Transforming the European energy system has become a major priority. Such transformation needs instruments that will optimize the process by answering policy questions such as:

1.How to achieve a low cost and low emissions energy mix?
2.How to achieve an energy mix that maximizes employment opportunities?
3.How to achieve an energy mix that has the maximum societal acceptance?
4.Which are the most competitive low carbon technologies in the medium and long term?
5.Where should new energy installations be best located?
6.In which R&D areas should a country invest?
7.How should a country develop energy interconnections with other European and non European countries?
8.How to improve energy efficiency?

The answering the above questions with the use of models, will provide guidance for the optimal route towards the achievement of a low carbon energy system. Within the present project the existing tools have been identified and their future development was suggested through specific roadmaps. In this way, the decision making process can be based on rational, reliable and transparent methodologies, taking in mind the complex interaction of technological and economic issues.

Finally, through the suggestion of common evaluation procedures for the available tools, through the suggestion of common methodologies , and through the roadmap for the development of commonly accepted tools in the future, the project will help in the harmonization of the decision making procedures and will provide the means for better understanding of energy issues across Member States.


Learning from the experiences of previous project results the consortium has developed a competitive exploitation and dissemination strategy for the whole project period, in order to achieve a measurable impact within target sectors.

Maximizing dissemination was achieved on the project deliverables which are knowledge sharing driven, not only from a content point of view, but also from a formal point of view thanks to the Public status of the deliverables. The fact that the project's outcome possibly sounds indifferent for the large audience was managed by accompanying each project deliverable with a newsletter and other channels of the European organizations. The consortium has used all their existing relations (all involved stakeholders of the energy projects where the beneficiaries and their partners participated) to increase the impact of its efforts.

Project website:

Related documents