Drive cycle identification and future energy-demand prediction are advantageous when developing hybrid propulsion systems. They are applicable to vehicles that are driven along the same route everyday such as buses, refuse-collecting vehicles (RCVs) or delivery vehicles. Drive cycle identification can be used to identify what power transients can be expected to prepare the power train to operate under these conditions. If the energy management algorithm of a hybrid vehicle can account for future energy demand, then it can be arranged in such a way that the non-fossil-fuel energy sources are fully depleted at the end of the drive cycle. Given that RCVs always drive in similar drive cycles, drive cycle has been modeled and its main characteristics parameterized. The model is separated into different drive cycles which are related to different power-consumption modes. In this paper, a new method to identify drive cycles and the energy left to finish a route is proposed. The drive cycle identification is based on artificial intelligence algorithms, which have been trained and tested with real data with an average efficiency in drive cycle identification of over 90%. The energy necessary to finish the route is based on vehicle energy models and statistical analysis. This method can be used in the daily management of fleet vehicles to replace fossil fuel by electric energy, as is demonstrated in the proposed examples.