Interpreting Energy Utilisation with Shapley Additive Explanations by Defining a Synthetic Data Generator for Plausible Charging Sessions of Electric Vehicles

Electric vehicles (EVs) are an effective solution for reducing reliance on non-renewable energy sources. However, the lack of charging infrastructure and concerns over their range are some of the biggest hurdles to adopting EVs. Charging infrastructure for EVs is, however, on the rise. Proper planning of charging stations vis-à-vis road networks and related points of interest such as transportation hubs, schools, shopping centres, etc., alongside such roads become vital to laying out a plan for such infrastructure, particularly for developing countries like India where EV adoption is relatively in a nascent stage. Synthetic datasets can help overcome these hurdles and promote EV adoption. This article presents a synthetic dataset mechanism for EV charging infrastructure planning, taking the Indian city of Berhampur, Odisha with its existiing EV charging infrastructure as a reference. The dataset includes information on the number of charging sessions for EVs, allocation to chargers in EVCS, reach time, charging start and end time, waiting time, total time spent at EVCS, total charged amount, energy used, and cost for charging. This information can help city planners and utilities identify the optimal locations for charging stations and plan for future charging infrastructure augmentation. The dataset can also be used to predict energy usage for the near future and identify the key factors affecting the planning with the help of Explainable AI (XAI) techniques. This information can help forecast the demand for charging services and optimize energy usage in the city. The article contributes to the EV charging behaviour and infrastructure planning and aims to promote broader EV adoption for future sustainable transportation.