Comperative analysis of air quality in European Union countries as a result of innovative clustering

Abstract

Purpose: The overarching aim of this article is to divide EU countries into classes characterised by different levels of ecology in the field of air quality using factors in the form of emission levels of harmful substances into the atmosphere. The data included in the study concern the year 2017 and come from OECD STAT. Approach/Methodology/Design: The classification of EU countries was carried out using the cluster analysis method as one from data mining methods. The results confirmed the hypothesis that the European Union countries can be divided into three distinct classes according to the criterion of environmental performance in the area of air quality. Findings: In terms of the conditions considered for the international air quality test, concentration 3 was the best, namely Austria, Belgium, Bulgaria, Croatia, Denmark and Finland, Greece, Lithuania, Latvia, Malta,Portugal, Romania, Slovakia, Slovenia, Sweden and Hungary. These are the countries that place the least burden on the atmosphere from harmful factors. The countries belonging to cluster 2, namely Cyprus, the Czech Republic, Estonia, the Netherlands, Ireland and Luxembourg, have also achieved high levels of harmful agents. It should be clearly indicated that the lowest level of emissions of harmful substances into the atmosphere was found in cluster 1, and more precisely in France, Spain, Germany, Poland, Italy, the UK and Wolf. Practical Implications: It turns out that the good practices and recommendations of any sustainable development service do not achieve their purpose in comparison with other EU countries. It seems necessary to tighten the ban policy by imposing high fines on the prohibition of the use of all prohibited objects, devices which have a significant impact on the deterioration of air quality. Originality/Value: There are many reports on the market in the scope discussed above, rich literature compilations. The proposed model may contribute to further development in order to implement the proposed breakdown to reflect actual results in a way that does not seem to hide the worst performing countries.