Queueing-based formulation model in a public transit network considering energy storage technology and demand charges

Battery Electric Buses (BEBs) are quickly emerging as a viable option for dramatically lowering transportationrelated emissions. Finding optimal locations of charging stations for BEBs can facilitate the development of electric transit networks that provide continuous services. Using fast charging along with Energy Storage (ES) charger technologies has also garnered attention to mitigate the peak power demand. A consensus regarding two types of chargers may result in more investment in the electric transit network, which resulted in considering installing fast chargers in designing an electric transit network. While this study aims to consider two types of fast- and ES-chargers, along with demand charges and waiting times at terminal stops simultaneously, to minimize the total costs of the electric transit network. The queuing model integrated with the Mixed-Integer Non-Linear Programming (MINLP) model is proposed to locate the optimal placements of charging stations in the electric transit network. To validate the model, a benchmark Mumford0 network is applied. Results indicated that the proposed charging location model could efficiently determine optimal locations of fast- and ES-chargers across the electric transit network. Also, the results confirmed that considering ES chargers could significantly save 16.6 % of the total costs of the Mumford0 transit network and reduce the total demand charge imposed on the network by the fast charger. More interestingly, by purchasing more chargers, transit agencies can decide on an acceptable waiting time that would be comfortable for BEBs arriving at terminal stops.

Automated summary

Battery Electric Buses (BEBs) are a viable option for reducing transportation emissions, and finding optimal locations for charging stations is crucial for the development of electric transit networks. Fast charging and Energy Storage (ES) chargers can help mitigate peak power demand, and considering the installation of fast and ES chargers can encourage investment in electric transit networks. This study proposes a queuing model integrated with a Mixed-Integer Non-Linear Programming (MINLP) model to determine optimal placements of charging stations in the electric transit network, taking into account demand charges and waiting times. The model is validated using a benchmark Mumford0 network, and the results show that it efficiently determines optimal locations for fast and ES chargers, resulting in cost savings and reduced demand charges.