Journal of the Bulgarian Geographical Society 46: 31-42, doi: 10.3897/jbgs.e89215
Clustering analysis of the light industry in Bulgaria
expand article infoAleksandra Ravnachka, Velimira Stoyanova
‡ National Institute of Geophysics, Geodesy and Geography - Bulgarian Academy of Sciences, Sofia, Bulgaria
Open Access

Light industry is one of the most important and priority industries in Bulgarian economy. It includes the production of textiles, clothing, and leather. Its development affects the state of the country’s overall economy. Despite the numerous studies that use GIS, in Bulgaria there have been no publications on the application of statistical analysis with the use of ArcGIS software. This study aims to apply Geographic cluster analysis using ArcGIS software to analyze the light industry in Bulgaria as of 2010, 2015, and 2020. The grouping of areas by selected indicators in the present study was performed with the Grouping Analysis tool. NO_SPATIAL_CONSTRAINT was selected for the Spatial Constraints parameter and FIND_SEED_LOCATIONS – for the Initialization Method. In this case, we used the K-Means algorithm to partition features into groups. That algorithm is one of the most popular and widely used clustering algorithms in GIS applications. The areas were grouped into 10 clusters. The selection of indicators on which the clustering procedure was based, is following the generally accepted indicators for assessing the state and importance of the food industry in the structure of the economy. The following indicators were used: output for 2010, 2015, and 2020; number of employees and export earnings as of 2010, 2015, and 2020, for each administrative-territorial unit. The spatial distribution of the population, in combination with the historical and the modern economic development of the settlements, forms the regional differences in the development of the light industry in the country. The cluster analysis of certain indicators for the assessment of the light industry at the NUTS 3 level as of 2010, 2015, and 2020, shows some changes in the spatial development trends of the industry. The cluster analysis shows that there are slight spatial differences in production at the NUTS 3 level, with large consumer centers and markets being the most important.

clothing production, grouping analysis, K-means clustering, leather production, statistical information, textile production