A Multicriteria Approach for Assessing the Impact of ICT on EU Sustainable Regional Policy

: As a global actor, the European Union (EU) plays a leading role in international e ﬀ orts to promote sustainable development globally. All sustainable objectives and targets need Information and Communication Technologies (ICTs) as key catalysts, since ICTs constitute tools of unprecedented power which help people to face the growing challenges of rising population, poverty, epidemics and climate change. Policy makers in the EU are increasingly putting ICTs into relations with sustainable regional development. This paper aims to study and assess the impact of ICT on the EU regional policy in terms of sustainable development by applying the multicriteria approach, PROMETHEE II, using the software Visual PROMETHEE. The criteria that were used in this research are the criteria that both the European Commission and member states deﬁne to assess the ICT implications of new EU legislation since 2010. The results revealed that the impact of ICT on EU sustainable regional policy has gotten stronger in the last two decades.


Introduction
Information and Communication Technology (ICT) advancements not only bring new opportunities, but also bring to light new risks for the achievement of sustainable development (SD) goals [1]. ICT proved to accelerate the worldwide socio-technological progress through knowledge transfer, marketing goods or services, network externalities and the development of cooperative relationships [2]. The contribution of ICT in addressing the major challenges of sustainable energy, climate change and sustainable development is highlighted by researchers, entrepreneurs, decision-makers and policy-makers [3]. The integration of ICT into social development and economic growth provides better opportunities for enhancing competitiveness and satisfying human needs [4,5].
The initiatives required to help ICT enable transformation to sustainable development and global competitiveness are as follows [6]: • Improve the four inter-related dimensions of ICT (the 4C Framework): Computers, Connectivity, Content, (human) Capacity.

•
The successful application ICT for sustainable development depends on the scalability and the sustainability.
-ICT constitutes just a means of achieving sustainable development. The main aim of the current policies and strategies for regional development is the flow of money from the rich nations to the poor nations, as well as the support to confront regional challenges through funding programs [30]. Aiming to create empowering partnerships among EU regions and EU member states in the context of these policies' implementation, the EU adopts some practices, such as monitoring and policy evaluation, in order to enhance the effectiveness of the policies [30]. The European Commission's involvement in regional development can be traced back to 1957 when the Treaty of Rome required the community to ensure "harmonious" development by reducing regional differences and the backwardness of less-favored regions [31].
According to the latest Eurobarometer regarding "Citizens' awareness and perceptions of EU regional policy" in 2017, almost 80% of the EU citizens were convinced that EU regional policy investments had a positive effect on their region or city, while almost half of the EU citizens supported the idea that EU regional investments should continue [32]. In order to be effective, a policy, such as regional policy, necessarily involves multiple partners operating at different spatial scales and different governance levels [33]. There are four structural funds: the European Regional Development Fund (ERDF), which is intended to finance large infrastructure projects and has the largest weight in the budget; the European Social Fund (ESF), which is the main financial instrument assisting the EU in realizing the strategy and the main objectives of its implementing policy; the European Agricultural Guidance and Guarantee Fund (EAGGF-Guidance Section), which accelerates the reformation of the agricultural structure; the Financial Instrument for Fisheries Guidance (FIFG), which is the specific fund for the reform of the structure of the fisheries sector [34]. The structural actions represent approximately one-third of whole the budget of the EU [34]. Infrastructure expenditures by are addressed to low-growth, low-employment and low-productivity regions [35].
It is widely accepted that EU regional policy has been an important contributing factor to the promotion of EU political regionalism and decentralization [36]. This is primarily as a result of the Funds of Partnership principle, taking into consideration that there are competent regional authorities and they get actively involved in the developing and implementation of regional aid [36]. Whereas a succession of income transfers may lead to a series of possible reductions in regional disparities, it should not be confused with the process of convergence on the regional level, which would be the successful implication of regional policies [37]. An emphasis on endogenous growth models has developed, and in connection with this, the intention to enhance the competitiveness among EU regions in contrast with simply getting involved in redistributive activities characterized by former regional policy interventions [38]. The European Regional Policy Research Consortium (EoRPA), originally launched in 1978, and was funded by government departments in Austria, Finland, France, Germany, Italy, the Netherlands, Norway, Poland, Sweden, Switzerland and the United Kingdom. It involves the monitoring and analysis of national regional policies in 30 European countries, and the study of the inter-relationships between EU Cohesion policy and EU Competition policy control of state aid [39].
The objective of this study is to investigate the impact of ICT on EU regional policy in the terms of sustainable development by applying the multicriteria approach, PROMETHEE II, using the software, Visual PROMETHEE. This approach was used for the assessment of the impact of ICT and for the ranking of the EU regional policies.

Materials and Methods
EU regional policies are retrieved from the official European Union website (www.europa.eu). The first step was to record the regulations, the directives, the decisions, the communications and other acts regarding the regional issues. EU sustainable regional policies form the alternatives. After the collection of the policies, a two-dimensional table was developed in order to find out the existence or lack of criteria that both the European Commission and member states have defined and proposed in order to assess the ICT implications of new EU legislation since 2010 [40]. These criteria constitute the variables X1, X2, ..., X12 (Table 1). Variable X1 refers to the requirement of the design of information rich Sustainability 2020, 12, 4869 4 of 13 processes by the legislation, while variable X2 refers to the requirement of the design of new business processes. Variable X3 represents the requirement of large amounts of data gathering in these processes and variable X4 represents the requirement of collaboration between ICT systems of multiple DGs or institutions/organizations. Variable X5 is about the fact that the legislation concerns ICT systems or that ICT is a supporting function of the legislation. The first five criteria/variables describe the level of dependence on ICT solutions of the EU regional policies and the weight of each is 1. The rest of the criteria/variables describe the levels of complexity of the ICT solutions and the weight of each is 0.83. The weights of the criteria were defined according to the method used by the European Commission to assess the ICT implications of EU legislation [40], giving the same importance to the level of dependence on the ICT implications and the level of complexity of the ICT implications. Variable X6 refers to whether the legislation requires new ICT solutions or existing applications can fulfill the requirements, while variable X7 refers to the existence of legacy systems which might hamper the implementation. Variable X8 concerns the imposition of authentication requirements by the legislation and variable X9 concerns the requirement of large amounts of data exchange between member states and/or the Commission. Variable X10 is about the required lead-time of the implementation (urgency), variable X11 is about the requirement of new interoperability specifications and variable X12 is about the imposition of high security requirements on the ICT solution by the initiative. The total amount of criteria achieved by each EU regional policy was also studied. Table 1. Criteria [41].

Category Variable Criteria
Dependence on the ICT solutions X 1 Does the legislation require the design of information rich processes?

X 2
Does the legislation require the design of new business processes?

X 3
Are large amounts of data gathering required in these processes?

X 4
Is collaboration between ICT systems of multiple DGs or institutions/organizations required?

X 5
Is the legislation concerning ICT systems or is ICT a supporting function of the legislation?

Complexity of the ICT solutions X 6
Does the legislation require new ICT solutions or can existing applications fulfill the requirements?

X 7
Are there any legacy systems which might hamper the implementation?

X 8
Does the legislation impose authentication requirements?

X 9
Is a large amount of data exchange between member states and/or the Commission required?

X 10
What is the required lead-time of the implementation (urgency)?

X 11
Are new interoperability specifications required?

X 12
Does the initiative impose high security requirements on the ICT solution?
Furthermore, the impact of ICT was assessed, and the EU regional policies were ranked using the multicriteria analysis, PROMETHEE II. PROMETHEE constitutes a prescriptive methodology Sustainability 2020, 12, 4869 5 of 13 which enables the decision-maker to rank the actions regarding his preferences [42]. Two rankings are calculated: the partial ranking is calculated mainly by undisputable preferences (PROMETHEE I), while the complete ranking, which is probably weaker, is obtained according to the decision-maker's requirements, too (PROMETHEE II) [42]. The PROMETHEE IV method solves a choice problematic for an infinite set of actions. It uses the same outranking relation, but the flows are defined on a compact subset of R" [43].
The PROMETHEE method includes four processes [44]: (1) Define the preferences: this function deals with the decision-maker's preference for an alternative x k with respect to another alternative x l regarding a criterion. (2) Calculate the preference index: this index is used to compare the alternatives in pairs, quantitatively taking into consideration all the defined criteria. (3) Construct valued outranking graph: outgoing and incoming flows are determined by means of relevant preference indices. (4) Rank alternatives according to the valued outranking graph: determination of the weights is an important step in most multi-criteria methods [45].
The method normalizes the weights of the criteria in order for their sum to be equal to 1.0 (100%) [44]. The PROMETHEE II method is described thoroughly in Brans and Mareschal (2005) [46] and in Andreopoulou et al. (2017) [47].
PROMETHEE provides the researcher with rankings of the alternatives and GAIA with a graphical representation of the decision problem [48]. The GAIA analysis is based on the uni-criterion net flows [42]. GAIA uses the principal components analysis (PCA), a well-known dimension-reduction technique for statistical data analysis [49].
PROMETHEE II methodology was selected in order to evaluate the impact of ICT on EU sustainable regional policies and to rank the policies because [50]: • there is now so much sensitivity of the estimated relation in small changes. • the results can be easily interpreted and discussed. • the use of the superiority relation is applied when the alternatives (sustainable regional policies) have to be ranked from the alternative with the highest score to the alternative with the lowest score. • the assessment and ranking process of complicated cases of sustainable regional policies is suitable for the application of PROMETHEE II methodology in the way that it seems to be closer to reality.

Results
The research through the official European Union website (www.europa.eu) resulted in the retrieval of 50 regional policies, which are presented in the Appendix A. In total, 19 out of the 50 EU regional policies have been established since 2010 (when the European Commission began to assess the ICT implications of EU legislation). Figure 1 presents the partial rankings of the 19 EU regional policies based on the computation of the two preference flows (Phi+ and Phi-). Figure 2 presents the complete rankings of the 19 EU regional policies, which are based on the total net preference flow (Phi). According to the results of PROMETHEE I and PROMETHEE II, COM(2012)19 "Waste Electrical and Electronic Equipment (WEEE)" is preferred to all other regional policies. COM(2012)19 has the highest score on Phi (0.5841), followed by Reg.2015/207 "The models for the progress report, submission of the information on a major project, the joint action plan, the implementation reports for the Investment for growth and jobs goal, the management declaration, the audit strategy, the audit opinion and the annual control report and the methodology for carrying out the cost-benefit analysis" and COM(2014)473 "Sixth report on economic, social and territorial cohesion: investment for jobs and growth", while COM(2015)118 "The Agreement on the European Economic Area" has the lowest score. The ranking of PROMETHEE I is confirmed by the ranking of PROMETHEE II.

Results
The research through the official European Union website (www.europa.eu) resulted in the retrieval of 50 regional policies. In total, 19 out of the 50 EU regional policies have been established since 2010 (when the European Commission began to assess the ICT implications of EU legislation). Figure 1 presents the partial rankings of the 19 EU regional policies based on the computation of the two preference flows (Phi+ and Phi-).

Figure 1.
Partial ranking of EU regional policies. Figure 1. Partial ranking of EU regional policies.
Sustainability 2020, 12, x FOR PEER REVIEW 6 of 13 Figure 2 presents the complete rankings of the 19 EU regional policies, which are based on the total net preference flow (Phi). According to the results of PROMETHEE I and PROMETHEE II, COM(2012)19 "Waste Electrical and Electronic Equipment (WEEE)" is preferred to all other regional policies. COM(2012)19 has the highest score on Phi (0.5841), followed by Reg.2015/207 "The models for the progress report, submission of the information on a major project, the joint action plan, the implementation reports for the Investment for growth and jobs goal, the management declaration, the audit strategy, the audit opinion and the annual control report and the methodology for carrying out the cost-benefit analysis" and COM(2014)473 "Sixth report on economic, social and territorial cohesion: investment for jobs and growth", while COM(2015)118 "The Agreement on the European Economic Area" has the lowest score. The ranking of PROMETHEE I is confirmed by the ranking of PROMETHEE II.  Figure 3 shows the PROMETHEE diamond, which depicts in a better way the two rankings (Phi+ and Phi-), while the vertical dimension represents the total net flow (Phi) by the complete ranking process. It can be observed that all action cones are located on the left axis (Phi+), which means that the total net flow of the regional policies is less than 1.

Figure 2.
Complete ranking of EU regional policies. Figure 3 shows the PROMETHEE diamond, which depicts in a better way the two rankings (Phi+ and Phi-), while the vertical dimension represents the total net flow (Phi) by the complete ranking process. It can be observed that all action cones are located on the left axis (Phi+), which means that the total net flow of the regional policies is less than 1. Table 2 shows the Phi scores of all the EU regional policies. The values calculated for the total net flows (Phi) present a large spectrum of values between +0.5841 and −0.4123, and that shows a great difference concerning "superiority" between the first and the last case in the ranking of EU regional policies according to the impact of ICT.  Table 2 shows the Phi scores of all the EU regional policies. The values calculated for the total net flows (Phi) present a large spectrum of values between +0.5841 and −0.4123, and that shows a great difference concerning "superiority" between the first and the last case in the ranking of EU regional policies according to the impact of ICT. Table 2. Preference flows.

Regional Policy Phi+
Phi In Figure 4, the GAIA plane is displayed. The red axis is the decision axis, which indicates the direction for the best solution according to the weight vectors on the GAIA plane. Because the direction of the decision axis is in the same direction as the variables X2 "The requirement of the design of new business processes" and X4 "The requirement of collaboration between ICT systems  In Figure 4, the GAIA plane is displayed. The red axis is the decision axis, which indicates the direction for the best solution according to the weight vectors on the GAIA plane. Because the direction of the decision axis is in the same direction as the variables X2 "The requirement of the design of new business processes" and X4 "The requirement of collaboration between ICT systems of multiple DGs or institutions/organizations", it can be expected that the PROMETHEE II ranked actions to be stronger on this variable and potentially weaker on variables X8 "The imposition of authentication requirements by the legislation" and X11 "The requirement of new interoperability specifications". COM(2010)553 "Regional Policy contributing to smart growth in Europe 2020" and Reg. of multiple DGs or institutions/organizations", it can be expected that the PROMETHEE II ranked actions to be stronger on this variable and potentially weaker on variables X8 "The imposition of authentication requirements by the legislation" and X11 "The requirement of new interoperability specifications". COM(2010)553 "Regional Policy contributing to smart growth in

Discussion
Research into the official EU website retrieved 50 regional policies, while 19 out of them have been established since 2010 (when the European Commission began to assess the ICT implications of EU legislation). The fulfillment of these 12 criteria, used by the European Commission, was studied in order to depict the impact of ICT on EU sustainable regional policy. Multicriteria decision analysis and GAIA analysis, are presented to identify the impact of ICT on EU sustainable regional policies.
The results and conclusions are summarized as follows: • PROMETHEE II analysis shows that ICT has the strongest impact on COM(2012)19 "Waste Electrical and Electronic Equipment (WEEE)", while Reg.2015/207 "The models for the progress report, submission of the information on a major project, the joint action plan, the implementation reports for the Investment for growth and jobs goal, the management declaration, the audit strategy, the audit opinion and the annual control report and the methodology for carrying out the cost-benefit analysis" is ranked in the second position. • The same analysis shows that ICT has the weakest impact on COM(2015)118 "The Agreement on the European Economic Area". • Most EU sustainable regional policies adopt ICT solutions, as most of them contribute positively to their total net flows (Phi). • The large spectrum of the values of the total net flows (Phi) shows a great difference concerning "superiority" between the first and the last EU regional policies, according to the impact of ICT.

Discussion
Research into the official EU website retrieved 50 regional policies, while 19 out of them have been established since 2010 (when the European Commission began to assess the ICT implications of EU legislation). The fulfillment of these 12 criteria, used by the European Commission, was studied in order to depict the impact of ICT on EU sustainable regional policy. Multicriteria decision analysis and GAIA analysis, are presented to identify the impact of ICT on EU sustainable regional policies.
The results and conclusions are summarized as follows: • PROMETHEE II analysis shows that ICT has the strongest impact on COM(2012)19 "Waste Electrical and Electronic Equipment (WEEE)", while Reg.2015/207 "The models for the progress report, submission of the information on a major project, the joint action plan, the implementation reports for the Investment for growth and jobs goal, the management declaration, the audit strategy, the audit opinion and the annual control report and the methodology for carrying out the cost-benefit analysis" is ranked in the second position.

•
The same analysis shows that ICT has the weakest impact on COM(2015)118 "The Agreement on the European Economic Area".

•
Most EU sustainable regional policies adopt ICT solutions, as most of them contribute positively to their total net flows (Phi).

•
The large spectrum of the values of the total net flows (Phi) shows a great difference concerning "superiority" between the first and the last EU regional policies, according to the impact of ICT. "The requirement of the design of new business processes" and "The requirement of collaboration between ICT systems of multiple DGs or institutions/organizations" are found to be the most robust criteria in the PROMETHEE II ranking.
• "The imposition of authentication requirements by the legislation" and "The requirement of new interoperability specifications" are found to be the weakest criteria.
As most EU sustainable regional policies present positive total net flows, this research confirms that the impact of ICT on EU sustainable regional policy has been getting stronger since 2010. The applied methodology constitutes an efficient planning tool at EU level for the policy makers for the assessment of sustainable regional policies, based on the impact of ICT. Furthermore, the findings of this research can be a supportive tool for the policy makers, as the superior EU regional policies can be used as benchmarks for future policies in terms of sustainable development. ICT adoption shows an enormous potential for accelerating the progress towards SDGs (Sustainable Development Goals), while at the same time it can improve the quality of life of people in fundamental ways [51]. However, it would be very interesting to apply different methods for multiple criteria decision-making because of some disadvantages of PROMETHEE method, such as the paradigm of the underlying method, the determination of the weights and the rank reversal problem [52].
Author Contributions: C.K. and Z.A. conceived of the presented idea. C.K. contributed to the design and the implementation of the research, to the analysis of the results and to the xriting of the manuscript. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.

Conflicts of Interest:
The authors declare no conflict of interest.  The models for the progress report, submission of the information on a major project, the joint action plan, the implementation reports for the Investment for growth and jobs goal, the management declaration, the audit strategy, the audit opinion and the annual control report and the methodology for carrying out the cost-benefit analysis.

COM(2014)473
Sixth report on economic, social and territorial cohesion: investment for jobs and growth.

COM(2009)295
Classification of certain goods in the Combined Nomenclature.

COM(2008)371
Fifth progress report on economic and social cohesion Growing regions, growing Europe.

COM(2007)273
Fourth progress report on economic and social cohesion Growing regions, growing Europe.

COM(2002)709
A framework for target-based tripartite contracts and agreements between the Community, the States and regional and local authorities.

COM(2003)811
Dialogue with associations of regional and local authorities. The interoperability of the rail system within the Community (Recast).