Data Driven Characterization and Predictive Classification of Energy Economy for Public Transport

The transformation towards sustainable mobility in public transportation sector requires thorough understanding of the use case, thus real world driving data and transparency in vehicle energy economy. Uncertainty about the energy demand leads to conservative design which lasts in inefficiency and high costs. Predicting the energy economy precisely upfront would significantly reduce costs and enhance fleet operations. Within this paper, we introduce a data driven approach for characterization and predictive classification of electric city buses by powerful machine learning algorithms. The presented framework facilitates the design and operation planning of alternative fleets in urban environment.