Initial location selection of electric vehicles charging infrastructure in urban city through clustering algorithm

Abstract

Transportation is one of the critical sectors worldwide, mainly based on fossil fuels, especially internal combustion engines. In a developing country, heightened dependence on fossil fuels affected energy sustainability issues, greenhouse gas emissions, and increasing state budget allocation towards fuel subsidies. Moreover, shifting to electric vehicles (EVs) with alternative energy, primely renewable energy sources, is considered a promising alternative to decreasing dependence on fossil fuel consumption. The availability of a sufficient EV charging station infrastructure is determined as an appropriate strategy and rudimentary requirement to optimize the growth of EV users, especially in urban cities. This study aims to utilize the k-mean algorithm’s clustering method to group and select a potential EV charging station location in Jakarta an urban city in Indonesia. This study proposed a method for advancing the layout location’s comprehensive suitability. An iterative procedure determines the most suitable value for K as centroids. The K value is evaluated by cluster silhouette coefficient scores to acquire the optimized numeral of clusters. The results show that 95 potential locations are divided into 19 different groups. The suggested initial EV charging station location was selected and validated by silhouette coefficient scores. This research also presents the maps of the initially selected locations and clustering.