1. Introduction
Evaluation of energy technologies and energy planning necessitates the participation of relevant stakeholders, from electricity producers and energy associations to environmental groups and local communities. Urban energy stakeholders include those who have legitimate responsibilities for energy projects (e.g., government authorities—national, regional, and local), those who support or oppose these initiatives (e.g., non-governmental organizations or NGOs, consumer associations, homeowner groups), and those who depend on it (e.g., energy users and customers). Each stakeholder group, however, has its own objectives, priorities, and preferences that should be taken into account during the process of energy technologies evaluation and planning.
One method for structuring and analyzing a multi-actor and multi-objective complexity is Multiple Criteria Analysis (MCA). MCA has been widely used for sustainable energy planning, as a useful tool in facilitating decision making among different stakeholder groups, in expanding the range of possible outcomes, and in assessing the performance of technologies against a set of evaluation criteria [
1,
2,
3].
However, a universal ranking of energy technologies, as has been attempted already [
4,
5,
6], would not be applicable in all cases and geographical contexts. Different geographical and jurisdictional levels would lead to selection of different criteria and therefore to evaluation from different perspectives with possibly different outcomes. Analysis of local energy stakeholders’ preferences at the European level, which to the best of our knowledge has not been previously performed, could provide useful insights for energy planning.
In many MCA applications, the direct inclusion of stakeholders is not considered. Often experts attempt to deduce stakeholders’ preferences instead of including them directly in the decision-making process. Most applications on energy issues focus on technical aspects, without involving stakeholders in the decision-making process in a constructive way [
2].
Our review of the energy planning literature showed that MCDA methods have been used extensively in Europe in the assessment of different energy options at different levels. MCDA approaches have been applied in the assessment of energy and low-carbon options mainly at the micro (project) level, but also at the meso (local/regional) and macro (national/international) levels.
Table 1 and
Table 2 summarize the studies that have applied MCDA approaches in meso and macro levels for assessing future and current energy options in Europe. Furthermore, the table provides information on the level of inclusion of stakeholders in the phase of criteria weighting. Detailed reviews of MCDA applications in energy planning have been done by different authors [
1,
2,
3,
7] In addition, analysis of the potentials and opportunities of using MCA in sustainability assessment [
8].
Table 1.
MCDA applications in energy planning in Europe at the local/regional level.
Table 1.
MCDA applications in energy planning in Europe at the local/regional level.
Level | Study | Thematic Area | MCA Methodology | Current vs. Future Energy Options | Weighting Method | Criteria Selection | Actors Involved in Weighting Process | Real Application |
---|
Chios Island, Greece | [9] | Renewable energy projects | Promethee II | Current | Direct weights | By researchers | Weight factors reflecting the analysts' previous experience | Proposed methodology |
Sardinia Island, Italy | [10] | Renewable energy technologies | Electre III | Current | SIMOS approach | By researchers | Three different scenarios by the researchers | Proposed methodology |
Salina Island, Italy | [11] | Wind energy plants | Naiade method | Current | Does not incorporate a traditional weighting technique | By researchers | Does not incorporate a traditional weighting technique | Yes |
Catalonia, Spain | [12] | Wind farm locations | Social multi-criteria evaluation | Current | Equal weights | By researchers and stakeholders | Equal weights were assigned | Yes |
Metropolitan Borough of Kirklees in Yorkshire, United Kingdom | [13] | Small-scale energy technology applications | MACBETH | Current | Direct allocation | By researchers | Five (5) professionals in the energy sector | Yes |
Norway (local case study) | [14] | Future energy-supply infrastructure | Equivalent attribute technique (EAT) | Future | Swing | By researchers | Six (6) professionals in the energy and research industry | Proposed methodology |
Crete, Greece | [15] | Sustainable energy planning | Promethee | Current | Direct allocation | Selected by researchers | Local authorities, local communities, potential investors, academic institutions, environmental groups, and government and European Union | Yes |
Urnasch, Switzerland | [16] | Future energy systems | Analytic heirarchy process | Future (2035) | AHP | By researchers and stakeholders | Energy consumers, experts and academics, and energy industry actors | Yes |
Thassos, Greece | [17] | Renewable energy sources | REGIME | Current | Direct allocation | By researchers | Criteria weights were determined based on the (1) combination of environmental, social, and economic characteristics of the technologies and (2) local and regional characteristics of the area under investigation. | Proposed methodology |
Crete, Greece | [18] | Strategic electricity generation planning | Delphi approach | Current | Direct allocation through Delphi | By researchers | A total of 30 experts (from the academe, national energy research centers, and power corporation). | Yes |
Table 2.
MCDA applications in energy planning in Europe at the national/international level.
Table 2.
MCDA applications in energy planning in Europe at the national/international level.
Level | Study | Thematic Area | MCA Methodology | Current vs. Future Energy Options | Weighting Method | Criteria Selection | Actors Involved in Weighting Process | Real Application |
---|
Bosnia and Herzegovina | [19] | Selection of energy system | ASPID | Current | Direct allocation | By researchers | Weighting factors were allocated to the different indicators by the researchers. | Yes |
Turkey | [20] | Future electricity resources | Promethee I and II | Future | Pairwise comparisons | By researchers | Criteria weighting was carried out by the researchers | Proposed methodology |
Greece | [21] | Alternative power generation scenarios | Promethee | Current | Direct allocation | By researchers | Four different weighting sets were used by researchers. | With attributes of real-world application |
United Kingdom | [22] | National energy policy | Simple multi-criteria evaluation | Future (2050) | Direct allocation | By stakeholders | Members of the general public by way of citizen panels and through (1) small group settings and (2) plenary for comparison of evaluations | Yes |
Austria | [23] | Energy scenarios | Promethee | Future (2020) | SIMOS | Selected by researchers and stakeholders | National case study stakeholders include government bodies, private firms, power distributers, NGOs and research institutes, while local case study stakeholders were energy experts, mayors, and citizens. | Yes |
Greece | [24] | Sustainable technological energy priorities | Linguistic variables | Future (2021) | Direct allocation | By stakeholders | Decision makers | Proposed methodology |
Spain | [25] | Selection of renewable energy project | VIKOR, Analytic Heirarchy Process | Current | Pairwise comparisons | By researchers | Three groups of stakeholders: government, banks, and development companies. | Proposed methodology |
Turkey | [26] | Energy planning | modified fuzzy TOPSIS | Current | Pairwise comparisons | By researchers | Three (3) energy planning experts | Proposed methodology |
Lithuania | [27] | Energy generation technologies | AHP and Additive Ratio Assessment (ARAS) method | Current | AHp | By experts | A group of 25 experts (managers and lawyers of energy enterprises, financial specialists, and scientists). | Yes |
Europe | [28,29,30] | Assessment of electricity supply options | Web-based MCDA | Future (2050) | Hierarchical weighting | By stakeholders | Stakeholders, ranging from energy suppliers and consumers to non-government organizations and government authorities. | Yes |
Based on our review, assessing future energy technologies while integrating and mapping local stakeholders’ perspectives at a wider scale (
i.e., European level) is lacking. The only European-wide MCA study that was conducted, which is the New Energy Externalities Development for Sustainability (NEEDS) project, applied a MCDA of future energy technologies in four countries, namely France, Germany, Italy, and Switzerland [
31,
32] for the year 2050. The MCA aimed to assess energy technologies, considering the varied national stakeholders’ preferences for the trade-offs between different criteria [
33].
Nevertheless, like in other research areas, a trend towards increased inclusion of stakeholders can be observed in energy research as well. The steps of criteria selection and weighting wherein stakeholders express subjective judgments are steps that could foster direct participation of stakeholders and inclusion of their preferences into the decision-making process [
7,
34]. However, the design and implementation of such interaction with stakeholders is considered a major challenge and should be carried out carefully [
28].
In addition, several studies have shown that well-articulated and preconceived preferences regarding unfamiliar and complex issues cannot apply [
35]. Instead, in these settings, respondents construct their preferences during the process of elicitation. Preference construction process should be considered when some of the decision elements are unfamiliar and where there are conflicts among the choices to be made [
36]. Energy planning and sustainability evaluation of energy systems are complex issues that also entail difficult decisions and trade off considerations. Moreover, preferences change under different contextual conditions [
37], while different methods (procedure) and different descriptions (framing) can give rise to systematically different responses [
38]. Hence, this indicates that respondents (and in our case local energy stakeholders) need a method to help them to articulate their preferences, and any attempt to derive their preferences should be based on an active procedure of preference construction [
36].
The combination of different methods during preferences’ elicitation and concluded, among other things, that combining different techniques could (a) provide a form of consistency testing and (b) lead to more reliable and acceptable elicitation of preferences has been investigated [
33]. The use of multiple methods to explore method invariance has been suggested as different methods can yield different results [
35]. Such inconsistencies are an opportunity to reflect on results from different framings of the issue at hand, whereas that opportunity is lost when a single method is used [
33]. Furthermore, as has been shown, some respondents may react negatively to the chosen approach, lessening acceptance of the process.
Moreover, it is argued that elicitation of preferences should be an iterative process, whereby the elicited values may have to be adjusted due to deviations from theoretical expectations or to an increased understanding of the problem and the context by the respondent [
39]. In addition, combined use of different methods and provision of technical support during the entire process results in minimization of potential biases, enhances appropriate use of the MCA methods, and facilitates confident expression of stakeholders’ preferences [
7,
33].
The current paper presents the development of a constructive criteria weighting approach that utilizes a consistency test of respondents’ preferences by incorporating different techniques to elicit local stakeholders’ preferences. At the same time, it builds an integrated sustainability criteria framework for the assessment of reference low-carbon energy technologies that would be deployed in 2030 in Europe. When respondents perform a series of choices during a weighting elicitation process, research has shown that it is very difficult to establish preference consistency [
39]. This article presents a constructive weighting process that tests respondents’ consistency and enables them to consistently express their preferences. This paper builds on a previous study by the authors [
7,
40] that incorporated stakeholders’ preferences in an energy and climate change policy context.
Moreover, this research identifies and analyzes the preferences of local energy stakeholders at the European level, which, to the best of our knowledge, has not been previously studied. The constructive weighting methodology and sustainability evaluation of low-carbon energy technologies was applied within the framework of Covenant CapaCITY, a European project that involved and supported different local stakeholders in the energy sector in Europe in developing Sustainable Energy Action Plans. The findings of this study, despite the limited survey sample, may advance the practice of a local stakeholder-driven process in the evaluation of low-carbon energy options in Europe and provide a framework to support group decision-making situations and policy design, two frequent and significant issues in the energy sector.
The paper is structured as follows:
Section 2 describes the main methodological aspects and data collection methods that were deployed for this research.
Section 3 reports the results of the pilot application of the constructive weighting methodology for eliciting local stakeholders’ preferences while also presenting the final ranking of low-carbon energy technologies. The final section includes the main implications of the research findings, future research directions, and concluding remarks.
3. Results and Empirical Analysis
3.1. Initial Ranking
Based on frequency count and percentages, the criteria that were considered to be of high importance by the weighting survey respondents were as follows: CO
2eq emissions, ecosystem damages, mortality and morbidity, accident fatalities, employment generation, levelized costs, resilience to climate change, and radioactive waste (
Figure 3).
Table 5 shows the results of the initial ranking and corresponding average ranking positions of the different evaluation criteria.
Figure 3.
Level of importance of the evaluation criteria.
Figure 3.
Level of importance of the evaluation criteria.
Table 5.
Initial ranking and corresponding average ranking positions of the evaluation criteria based on respondents’ preferences.
Table 5.
Initial ranking and corresponding average ranking positions of the evaluation criteria based on respondents’ preferences.
Initial Ranking | Criteria | Average Ranking Position |
---|
1 | CO2eq emissions | 3.50 |
2 | Levelized costs | 5.06 |
3 | Ecosystem damages | 5.94 |
4 | Accident fatalities | 6.75 |
5 | Mortality and morbidity | 7.19 |
6 | Employment generation | 7.38 |
7 | Radioactive waste | 9.38 |
8 | Fuel use | 9.63 |
9 | Resilience to climate change | 9.75 |
10 | Energy cost sensitivity to fuel price fluctuation | 10.50 |
11 | Stability of energy generation | 10.88 |
12 | Waste disposal (infrastructure) | 11.06 |
13 | Innovative ability | 11.13 |
14 | Technological maturity | 12.19 |
15 | Peak load response | 12.69 |
16 | Noise | 14.25 |
17 | Land use requirement | 14.50 |
18 | Market size (potential export) | 14.69 |
19 | Level of public resistance/opposition | 15.13 |
20 | Market concentration on supply | 15.38 |
21 | Market size (domestic) | 15.81 |
22 | Aesthetic/functional impact | 17.63 |
The initial ranking shows that CO
2eq emissions is the most preferred criterion, with an average ranking position of 3.5 (
Table 5). This is followed by levelized costs, ecosystem damages, accident fatalities, mortality and morbidity, employment generation, radioactive waste, fuel use, resilience to climate change, and energy cost sensitivity to fuel price fluctuation. Rounding off the list are stability of energy generation, innovative ability, waste disposal (infrastructure), technological maturity, peak load response, noise, land use requirement, market size (potential export), level of public resistance/opposition, market concentration on supply, market size (domestic), and lastly, aesthetic/functional impact.
3.2. Pairwise Comparisons Results
The initial ranking provided the basis for the consistency check. As such, the results of the initial ranking were compared with the results (final ranking) of the series of pairwise comparisons.
Table 6 presents the consistency levels that respondents achieved.
Table 6.
Respondents’ consistency levels.
Table 6.
Respondents’ consistency levels.
| Values | Number of Respondents |
---|
Low | <0.5 | 5 |
Moderate | 0.5–0.7 | 4 |
High | >0.7 | 7 |
As there were some respondents who achieved low and moderate consistency, where few of them were in favor of the initial ranking, the ranking outcome of the pairwise comparisons was not considered. The large number of pairwise comparisons in these cases probably imposed a high cognitive burden on the respondents who proved inconsistent. Therefore, in these cases, weights were determined based on respondents’ initial ranking by applying Formula (2). The constructive process that was integrated in the weighting method on one hand tested the consistency of stakeholders’ preferences and on the other hand “forced” the stakeholders to rethink, revise their initial preferences, and better evaluate the issue of criteria importance.
Based on the results of the approach, CO
2eq emissions topped the list, with an average weighting score of 0.083. Levelized costs, ecosystem damages, mortality and morbidity, and resilience to climate change were on the list of top five preferred criteria.
Figure 4 illustrates the final criteria weights based on stakeholders’ preferences. These results, though, should be further tested through a larger sample, whereby trends and patterns of local stakeholders’ preferences can be revealed.
3.3. Stakeholder Groups Preferences
From the distribution of weighting scores in
Figure 5, all three groups of local stakeholders expressed high preferences for CO
2eq emissions, levelized costs, ecosystem damage, and resilience to climate change. CO
2eq emissions was the most preferred criterion by both energy industry actors and technical professionals, while this ranked fifth among public authorities.
Figure 5 also shows the convergence and divergence of preferences among the three different local stakeholder groups.
It could be observed that public authorities gave more importance to ecosystem damage, which ranked second on the list. Moreover, public authorities expressed high preferences for social criteria. Mortality and morbidity was considered as the number one criterion, while accident fatalities ranked third.
Energy industry experts also showed high preference for mortality and morbidity. However, this criterion was not given much importance by technical professionals. Accident fatalities, however, was ranked eighth among technical professionals and twelfth among energy industry actors. Meanwhile, technical professionals had expressed high preferences for fuel use, which ranked second among this stakeholder group. It could also be observed that, compared to public authorities and energy industry experts, technical professionals expressed a preference for certain energy and technological criteria.
Technological maturity and market size—both domestic and potential exports, for example, received more weight from technical professionals compared to the other stakeholder groups. It could also be observed that public authorities, compared to energy industry experts and technical professionals, gave relatively low weight to certain energy and technological criteria, such as market size—domestic and potential export, stability of energy generation, and peak load response.
Also, energy public professionals and technical professionals gave the same weight to radioactive waste, while energy industry experts gave a relatively lower weight to this criterion. Technical professionals also gave relatively lower weight to social criteria, such as mortality and morbidity and accident fatalities, compared to the other two stakeholder groups. Interestingly, energy and industry actors gave relatively higher weight to the level of public resistance/opposition and aesthetic/functional impact compared to the other groups.
Figure 4.
Criteria weights including average, max, and min values.
Figure 4.
Criteria weights including average, max, and min values.
Figure 5.
Criteria weights according to the three stakeholder groups.
Figure 5.
Criteria weights according to the three stakeholder groups.
3.4. Clustering Priorities
Although the above analysis of weights and stakeholders’ priorities is insightful, this data can be analyzed further to see the level of homogeneity of stakeholder groups and if there are any identifiable “priority bundles” or groupings of criteria that respondents tend to prioritize. It can further be examined to see if certain “types” of stakeholders tend to give more weight to certain criteria categories. Cluster analysis was conducted on the survey data with these objectives in mind. Hierarchical agglomerative clustering methods were employed. First, each individual stakeholder was considered as an individual cluster, and then close pairs of clusters were merged using Ward’s method for clustering and the squared Euclidean distance to measure the distance between different observations. The resulting dendrogram suggests a three-cluster solution (
Table 7). Stakeholders (respondents) in each cluster have weighted similar groups of criteria and utilize similar priorities when evaluating low-carbon energy options (see
Figure 6). According to cluster weights and their average values, three “priority clusters” were created:
- (1)
Energy market priorities: Stakeholders (respondents) in this cluster gave higher priority (weight) to energy and technological criteria.
- (2)
Environmental priorities: Stakeholders (respondents) in this cluster gave higher priority (weight) to most of the environmental criteria.
- (3)
Socioeconomic priorities: Stakeholders (respondents) in this cluster gave higher priority (weight) to most of the social and economic criteria.
Table 7.
Cluster analysis results.
Table 7.
Cluster analysis results.
Criteria Categories | Criteria | Cluster 1: Energy Market Priorities | Cluster 2: Environmental Priorities | Cluster 3: Socio-economic Priorities |
---|
Maximum | Average | Minimum | Maximum | Average | Minimum | Maximum | Average | Minimum |
---|
Economic | Levelized costs | 0.09 | 0.07 | 0.00 | 0.09 | 0.06 | 0.03 | 0.16 | 0.09 | 0.06 |
Employment generation | 0.07 | 0.04 | 0.01 | 0.08 | 0.06 | 0.04 | 0.08 | 0.05 | 0.02 |
Environmental | CO2eq emissions | 0.13 | 0.07 | 0.00 | 0.16 | 0.10 | 0.05 | 0.09 | 0.06 | 0.03 |
Resilience to climate change | 0.13 | 0.06 | 0.00 | 0.14 | 0.09 | 0.02 | 0.08 | 0.04 | 0.00 |
Noise | 0.08 | 0.05 | 0.01 | 0.08 | 0.04 | 0.01 | 0.04 | 0.02 | 0.00 |
Radioactive waste | 0.06 | 0.03 | 0.00 | 0.14 | 0.08 | 0.02 | 0.06 | 0.04 | 0.02 |
Waste disposal (infrastructure) | 0.04 | 0.03 | 0.00 | 0.10 | 0.06 | 0.03 | 0.07 | 0.03 | 0.01 |
Ecosystem damages | 0.06 | 0.03 | 0.00 | 0.15 | 0.09 | 0.05 | 0.14 | 0.08 | 0.05 |
Land use requirement | 0.06 | 0.03 | 0.01 | 0.07 | 0.04 | 0.02 | 0.03 | 0.02 | 0.01 |
Fuel use | 0.18 | 0.08 | 0.02 | 0.06 | 0.04 | 0.02 | 0.07 | 0.03 | 0.01 |
Social | Level of public resistance | 0.04 | 0.03 | 0.02 | 0.05 | 0.03 | 0.01 | 0.07 | 0.04 | 0.01 |
Aesthetic/ functional impact | 0.06 | 0.03 | 0.02 | 0.04 | 0.02 | 0.00 | 0.04 | 0.02 | 0.00 |
Mortality and morbidity | 0.09 | 0.04 | 0.02 | 0.11 | 0.06 | 0.02 | 0.20 | 0.11 | 0.00 |
Accident fatalities | 0.08 | 0.05 | 0.02 | 0.10 | 0.06 | 0.01 | 0.10 | 0.06 | 0.03 |
Energy | Energy cost sensitivity to fuel price fluctuation | 0.07 | 0.05 | 0.02 | 0.06 | 0.03 | 0.01 | 0.06 | 0.04 | 0.02 |
Stability of energy generation | 0.08 | 0.05 | 0.03 | 0.04 | 0.02 | 0.01 | 0.08 | 0.04 | 0.01 |
Peak load response | 0.11 | 0.05 | 0.02 | 0.04 | 0.02 | 0.00 | 0.05 | 0.03 | 0.01 |
Energy | Market concentration on supply | 0.10 | 0.04 | 0.02 | 0.02 | 0.01 | 0.00 | 0.05 | 0.03 | 0.01 |
Technological | Technological maturity | 0.16 | 0.06 | 0.02 | 0.09 | 0.03 | 0.00 | 0.06 | 0.04 | 0.02 |
Market size (domestic) | 0.08 | 0.04 | 0.01 | 0.03 | 0.01 | 0.00 | 0.06 | 0.02 | 0.01 |
Market size (potential export) | 0.12 | 0.04 | 0.01 | 0.03 | 0.01 | 0.00 | 0.07 | 0.03 | 0.01 |
Innovative ability | 0.11 | 0.05 | 0.01 | 0.05 | 0.02 | 0.00 | 0.07 | 0.05 | 0.02 |
Regarding the homogeneity of stakeholder groups, the following observations can be drawn from the cluster analysis (see also
Table 8). The highest homogeneity was observed in the Public Authorities group, where four out of five respondents belong in cluster 2, which places more emphasis on environmental priorities. Technical experts proved to be relatively homogenous: three out of five respondents belong in cluster 1, which places more emphasis on energy market priorities; the other two respondents belong in cluster 2, which emphasizes environmental priorities. Energy industry actors also proved to be relatively homogenous, as three out of five respondents belong in cluster 3, which emphasizes socioeconomic priorities, whereas the other two belong in cluster 1, which gives priority to energy market considerations.
Table 8.
Type of stakeholders in each cluster (priorities).
Table 8.
Type of stakeholders in each cluster (priorities).
| Technical Experts | Energy Industry Actors | Public Authorities | Total |
---|
Energy Market Priorities | 3 | 2 | 0 | 5 |
Environmental Priorities | 2 | 0 | 4 | 6 |
Socioeconomic Priorities | 0 | 3 | 1 | 4 |
Total | 5 | 5 | 5 | 15 |
3.5. Evaluation of Energy Technologies
The evaluation of low-carbon energy technologies was conducted, and it was found that the highest ranked low-carbon energy technology is wind off-shore (0.79), followed by solar PVs (0.78), hydropower (0.74), wind on-shore (0.73), GTCC (0.58), GTCC with CCS (0.57), EPR (0.57), biomass (0.56), IGCC with CCS (0.53), and IGCC (0.45).
Figure 7 shows the final scores of each low-carbon energy technology, illustrating the contribution of each evaluation criterion to the final score. As can be observed from
Figure 7, technologies with high scores at the most important criteria, weighted by the stakeholders, in principle achieve higher overall final scores.
Figure 6.
Cluster analysis—clusters’ weighted averages.
Figure 6.
Cluster analysis—clusters’ weighted averages.
Figure 7.
Final scores of the low-carbon energy technologies and contribution of evaluation criteria based on local stakeholders’ preferences.
Figure 7.
Final scores of the low-carbon energy technologies and contribution of evaluation criteria based on local stakeholders’ preferences.
It could be observed (
Figure 8) that among the three stakeholder groups, wind off-shore is the highest ranked low-carbon energy technology. Solar PV is the second-ranked technology for energy industry actors and technical professionals, while public authorities favored hydropower. Solar PV is the third-ranked technology among public authorities, while energy industry actors favored hydropower. Meanwhile, wind on-shore is the third-ranked technology among technical professionals, while public authorities and energy industry actors ranked it fourth. It could be observed from the rankings among the three local stakeholder groups that renewable energy technologies outrank other technologies, such as fossil-fuel based ones (e.g., IGCC and GTCC) and nuclear technology (EPR).
Figure 8.
Overall weighted scores of low-carbon energy technologies per stakeholder group.
Figure 8.
Overall weighted scores of low-carbon energy technologies per stakeholder group.
4. Discussion
4.1. Criteria Weighting
As the initial ranking provided the basis for the consistency check, the results of the pairwise comparisons were checked in order to assure their consistency and reliability. By applying different ranking and weighting techniques, an opportunity for a consistency check was established to enhance the consistency of stakeholders’ preferences. Most of the time, this is neglected [
33]. By using different methods we were able to detect inconsistencies by comparing the different ranking results. Such inconsistencies are an opportunity to reflect on different framings of the issue at hand [
33]. By using a single method, such an opportunity is lost. Furthermore, MCA practitioners often apply a “one-size-fits-all” approach even though different methods work better for some people and situations than for others [
33].
It is important to note that 31% of the respondents achieved low consistencies between their initial and final rankings. As has also been stated by various authors [
7,
39,
62], the study conveyed that the difference in consistency between weighting methods could be related to the large number of criteria for comparison, particularly common in the case of pairwise comparisons. This research study involved 22 pairs of criteria for comparison, which resulted in a high cognitive burden on the respondents. Hence, with the large number of pairs for comparison, inconsistencies inevitably arose. This was expected as during prescriptive decision analysis processes, perceptions change and evolve, and the representation of these perceptions are not static [
39]. The respondents were then asked to modify their preferences should their weighting scores not reach the consistency threshold value. However, having to repeat the pairwise comparisons could have been a challenge for some of the respondents since this would have required additional time.
Furthermore, it could be observed that due to the cognitive demands as well as time constraints, the respondents were more comfortable with providing the ranking order directly for a list of criteria rather than selecting which criteria are relatively more important for each pairwise comparison. As other authors [
59,
60] suggest for the elicitation of weights, ranking methods using surrogate weights proved to be less cognitively demanding. In a previous study by the main author [
7], a sample of individual stakeholders and experts in the climate and energy policy field has expressed satisfaction as well as approval for combining ranking and pairwise comparisons as approaches in weighting energy and climate criteria. The study showed that the initial ranking facilitated a gradual approach to the evaluation problem. The pairwise comparisons, on the other hand, enabled a more accurate expression of the respondents’ preferences. The number of criteria (14) in that study was significantly fewer than the number of criteria (22) selected to be assessed in this study.
Cognitive limit is one of the challenges in stakeholders’ preference elicitation. In a decision problem that involves a small set of alternatives and criteria, most people can make their selection intuitively. However, with a large set of alternatives and criteria, relying on intuition and/or experience seems inadequate and thus needs further support. The conclusions correspond to the additional challenge of the mix of qualitative and quantitative indicators as well as of preferences that are often times irregular, non-sequential, and with threshold values [
28]. The computerized interaction was considered important in helping stakeholders to construct their preferences. This provided stakeholders with support in analyzing their desired objectives in relation to the outcomes of the elicitation process. Practical techniques for elicitation are to a great extent a matter of balancing the quality of elicitation results with the time available and cognitive burden on the respondents for eliciting all the required information [
39].
4.2. Stakeholders’ Preferences on Sustainability Evaluation Criteria
Local stakeholders, in general, expressed high preferences for CO2eq emissions, levelized costs, ecosystem damages, employment generation, resilience to climate change, fuel use, and waste disposal, which shows implied responsibility towards local benefits and negative externalities. Mortality and morbidity, accident fatalities, and radioactive waste were also high priorities for the respondents, which shows how local stakeholders value the welfare of the public, including workers, during project installation and operation. The potential impact of energy technologies on human health and safety is considered a priority.
Local stakeholders, and society in general, are still concerned about radioactive waste because of its potential to cause catastrophic accidents—whether likely or unlikely—or be used in terrorist attacks. In the aftermath of the Fukushima nuclear disaster in Japan in 2011, radioactive waste and nuclear safety remain controversial topics. Aesthetic/functional impact was not a significant preference among the local stakeholders. Although debate is inevitable regarding the aesthetics of current infrastructure of specific renewable energy technologies (e.g., wind and solar), mechanisms are available for the deployment of these technologies in unobtrusive ways [
63].
Public authorities prioritize public health protection and safety—and in general, certain social criteria—as proven by their preference for mortality and morbidity and accident fatalities. Public authorities also give significant priority to ecosystem damage, CO2eq emissions, and levelized costs, which reflects their concern for local environmental protection as well as economic outlay.
In spite of sharing similar preferences with public authorities and energy industry experts, technical professionals have a unique high preference for fuel use. This research study also concludes that technical professionals, when compared to other stakeholder groups, have higher preferences for certain energy and technological criteria. On the other hand, public authorities give the least priority to certain energy and technological criteria, while technical professionals have the lowest preferences for certain social criteria. However, this sample of the stakeholders does not allow for generalization of the results and indicates the need to apply this methodology in a larger sample of different local stakeholder groups around Europe.
With regard to the homogeneity of stakeholders’ groups and according to the cluster analysis that was conducted, public authorities proved to be the most homogeneous group, with a clear focus on environmental priorities. This is considered reasonable since the composition of the group of respondents was to a large extent linked to the Covenant CapaCITY European project, which has a clear focus on environmental aspects. Technical experts proved to be a relatively homogenous group, with emphasis on energy market and environmental priorities. The energy industry actors group also proved to be relatively homogenous, as most of the respondents emphasized socioeconomic priorities, whereas few more emphasized energy market priorities. These results provide the first insights on local stakeholders’ preferences typology. This typology and results need to be further explored on a larger scale in order to be able to generalize and come up with valuable recommendations for energy policy making.
4.3. Ranking of Energy Technologies
This research concludes that wind off-shore, solar PV, hydropower, wind on-shore, and GTCC are the low-carbon energy technologies that rank highest while considering the preferences of local stakeholders. On the other hand, IGCC with CCS and IGCC were the least significant low-carbon energy technologies among all three stakeholder groups.
The results of the NEEDS project [
30] also showed high preferences for renewables, such as solar, wind, and biomass technologies. Centralized gas options (e.g., combined cycle and combined heat and power CHP) as well as nuclear technologies were the mid-performing group of technologies, while coal and lignite technologies were considered the worst performers. In Turkey [
20], wind power proved to be the preferred option in their ranking of alternative energy sources. Wind was also the highest ranked alternative, followed by biomass and PV, in an MCA in Greece [
17].
The results of the study also show how certain technologies (e.g., renewables) that rank relatively low in cost-based assessments are otherwise most preferred and highly ranked if multiple sustainability criteria are considered in the assessment. One can surmise that economic costs certainly play a role in decision making, regardless of stakeholders’ propensity for choosing other sustainability criteria. As demonstrated in the results of the study, costs matter, but only up to a certain point. Other sustainability criteria, such as social and environmental ones, should also drive the assessment process.
4.4. Implications for Sustainable Energy Policy
As for low-carbon energy policy, it can be concluded that based on the overall preferences of stakeholders, there should be a focus on policies that enable the local deployment of renewable energy technologies reflecting preferred local priorities, such as levelized costs, ecosystem damages, and employment generation. Moreover, key differences regarding local stakeholder preferences could be highlighted during local low-carbon energy planning. Within the decision-making context, relevant stakeholders and decision makers would have informed opinions about the value judgments of local stakeholders that need to be taken into account in the process of developing low-carbon energy policies and sustainable energy action plans.
5. Conclusions
The constructive weighting methodology applied in this study allows for a thorough process for eliciting weighting preferences. The methodology requires survey respondents to be consistent in their preferences. Moreover, the use of different techniques enhances the reliability of the results as respondents had the opportunity to check and revise their preferences. The low consistencies of respondents’ preferences could be attributed to the large number of criteria involved and the cognitive burden this imposes to respondents. However, the demonstration of the constructive weighting methodology shows great potential for better decision making, supporting stakeholders’ efforts to gradually construct their preferences.
Overall, the research study was able to map, albeit in a limited manner, the preferences of local energy stakeholders. Based on the elicited preferences, the low-carbon energy technologies that best meet the evaluation criteria prioritized by local energy stakeholders were assessed. This study shows which sustainability (i.e., economic, environmental, social, energy, and technological) criteria local energy stakeholders prioritize during the assessment of low-carbon technologies. Furthermore, considering stakeholders’ priorities and preferences of sustainability criteria, a final ranking for low-carbon energy technologies is determined.
The results show clearly that there are converging stakeholders’ views on many aspects of sustainability assessment of low-carbon energy technologies, which implies the possibility of reaching a consensus between different local stakeholders in energy planning and finding the right balance between economic, environmental, social, energy, and technological considerations. The fact that renewable energies (RE) proved to perform best within the proposed integrated assessment framework leads to the conclusion that European and national policies should further enable RE deployment, not only to achieve low-carbon development objectives but also to meet local stakeholders’ priorities.
In this study, a constructive weighting methodology was applied to elicit European local stakeholders’ preferences on the evaluation criteria of future low-carbon energy technologies. However, this research study merited a small number of respondents based on a European project that aims at supporting local governments in conducting Sustainable Energy Action Plans. This research study mapped three broad categories, namely public authorities, energy industry actors, and technical professionals. It would be useful to map the preferences of distinct local stakeholder groups that apply within a larger local energy context in Europe. As such, there is a need for further application of the constructive weighting methodology to a large number of local energy stakeholders at the European level and, even better, to stakeholders groups’ decision-making process.
The constructive weighting methodology for this study can also be applied in a group decision context, wherein local stakeholders and decision makers meet face-to-face, e.g., workshops or consultation meetings. Furthermore, this weighting methodology could be carried out through an online process of interaction, e.g., webinar. Furthermore, different weighting methods could be tested to compare any differences in the results. Also, by applying different weighting methods, researchers can also examine the level of consistency of stakeholder preferences and how this is affected by the type of weighting methodology and framing.
Lastly, in situations where decision makers have to engage in the development of low-carbon energy strategies and sustainable energy action plans, local stakeholders’ preferences can be mapped out by applying this methodology. This is crucial also for the identification of potential conflicts and resolution of actual ones in order to reach consensus on the development of local sustainable energy strategies.