On the Spatial Dependencies of Human Mobility and Urban Energy Consumption

Urban areas are responsible for consuming up to 80% of the energy produced worldwide, mainly as a result of human activities. Due to the constantly increasing world population and the shift of this population into cities, over 60% of the world population is projected to reside in urban areas by 2030 and the corresponding increase in human activities will lead to a tremendous increase in energy consumption. Current approaches to energy consumption take a narrow sectoral approach and overlook the effects of individuals’ collective consumption as they visit different functional locations in a city in the course of their daily lives. They, therefore, underestimate consumption measures for exclusive vs. shared energy resources and fail to identify patterns of urban energy consumption with respect to consumers. Unreliable predictions and poor management decisions regarding future patterns of energy consumption and demand may thus lead to enormous waste in energy distribution and infrastructure investment. This paper explores the potential for developing valuable insights into energy consumption patterns in urban areas based on human activities inferred from the mobility behavior of urban populations. Through a study in Greater London covering the month of August, 2014, we analyzed 2,367,967 positional records from a location-based online social network (Twitter), and energy consumption (i.e., electricity and gas) data across 983 areas. A spatial autocorrelation analysis revealed clustering patterns for both electricity and gas consumption, as well as human mobility. Further, our spatial regression models indicate that human mobility can account for much of the distribution of energy consumption in urban environments and can be used to predict energy consumption patterns across urban areas. These results suggest data-driven approaches based on combining the mobility behavior of urban populations with geographical data including energy consumption and point of interest (POI) information can lead to further energy discoveries in urban functional regions. These findings will be of value to business practitioners, policy-makers, and research communities, enhancing their future efforts and enabling them to deal with overlooked or poorly specified aspects of urban energy consumption.