Unveiling the factors affecting final electricity consumption: does the regional component matter?

Abstract

Purpose The purpose of this paper is to explore the developments in final electricity consumption, estimate the portions of changes that can be attributed to national, sectoral or regional factors, and to investigate determinants of the regional component (RC) in Croatia at the subnational level in the period 2001-2013. Design/methodology/approach In the first stage, the dynamic shift-share method is used to decompose final electricity consumption, and then, in the second stage, the panel population-averaged logit model is conducted to find the main determinants of the extracted RC. Findings The results show that both the sectoral factor and the regional factor are responsible for an increase in electricity consumption over the period considered, whereby the regional specificities had a larger impact in general. Thereby, the most developed regions, including the tourism-oriented ones, exhibited the largest average increase in electricity consumption mainly due to positive effects of the regional-specific factors, while the negative effects of these factors were mainly responsible for low average rates of changes in electricity consumption in less developed regions. Practical implications The results suggest that regional-specific energy conservation programs might be more effective in improving energy efficiency than the sector-oriented ones, as well as that socio-economic and contextual determinants matter when it comes to the probability of having a positive regional effect on the electricity consumption rate. Originality/value The paper investigated the determinants of the extracted RC which has not yet been addressed in the energy economics literature.

Big data driven multi-tier architecture for electric mobility as a service in smart cities

“…In the mobility sector, these generated data sets are large enough, usually dynamic, unstructured, in different varieties, thus difficult for traditional database infrastructures to store, manage and process such data (Badii et al, 2017). This results in the increasing need for efficient and practical big data analytics tools to process the data by converting, analyzing and using the data toward realizing beneficial information to improve mobility services (Al-Jaroodi and Mohamed, 2018;Borozan and Pekanov Starcevic, 2018). Similarly, data from transport companies, public administrations and municipalities can be coarsely gathered as open data, which are data that can be used freely by anyone (Badii et al, 2017).…”

Predicting customer satisfaction for distribution companies using machine learning

“…Even in countries where the residential customers do not have access to an open energy market (e.g., in Brazil, although changes are under discussion) (Ministério de Minas e Energia, 2017), there is also growing concern about the quality of service provided and consumer satisfaction. However, it is hard to find a cause-and-effect relationship between service quality, customer satisfaction, energy consumption, regional components (Borozan and Starcebic, 2018) and the financial performance of companies operating in these markets. It is necessary a broader view of performance in a more economical than financial perspective, i.e.…”