Data-Driven Simulations for E-Mobility: Performance Across Italian Cities

As mobility evolves, Electric Vehicles (EVs) are becoming increasingly prevalent, offering new challenges and opportunities for understanding user behavior and optimizing charging strategies. This thesis presents the enhancement and evaluation of a Battery Electric Vehicle (BEV) simulator, along with an analysis of the factors behind unsatisfied trips for specific EV models under predefined charging policies. The study uses real-world driving data on ICE cars provided by UnipolTech. A previous version of the simulator developed by the research team was enhanced in terms of efficiency, scalability, realism, and flexibility, and models the behaviors of diverse charging profiles, specifying constraints based on time of the day, day of the week, charging durations, state-of-charge (SoC) thresholds, and charger preferences (AC or DC); integrating 50 EV models with different battery capacities, energy consumption rates and charging capabilities. From this catalog of EV models, five representative cars were selected for detailed analysis, ranging from small-battery EVs like the Dacia Spring to high-autonomy vehicles such as the Audi A6 Sportsback e-tron. A central focus of the study is the categorization and analysis of unsatisfied trips; those that could not be completed due to insufficient battery charge or were inhibited from recharging due to restrictions from predefined charging policies. For this study, nine different charging policies were simulated and analyzed, ranging from highly flexible to time-restricted approaches. Anonymized trip data were obtained from five Italian provinces: Milano, Asti, Grosseto, Sassari, and Trieste; selected to represent diverse geographical, demographic, and territorial contexts. These cities vary in population density and area, seeking to capture variations in driving behavior influenced by urban size, road networks, and typical trip patterns. Over 90% of trips are shorter than 35 km, while seasonal trends highlight longer trips during summer vacations and specific holidays. The results of the simulations indicate that flexible charging policies consistently achieve the highest satisfaction rates, often exceeding 93%, whereas restrictive policies lead to a greater proportion of unsatisfied trips. EV adoption and trip satisfaction depend not only on vehicle performance but also on aligning charging policies with specific user behavior conditions. Cross-city analysis showed that, despite differences in territory and population density, average trip distances, including typical maximum distances and average parking times per user, fall within the capabilities of small-battery EVs. With a flexible charging policy, Milano could reach up to 85% of satisfied users (being the least adaptable), while Sassari emerges as the most adaptable city, achieving 98% satisfaction, mostly with low-range EVs. However, under stricter policies, the forecast changes significantly: overall satisfaction rates drop, with Sassari still leading at 46.5%, followed by Milano at 39.3%. In this scenario, fulfilling user needs requires higher-range EVs compared to the flexible policy case. The study suggests that, with well-designed policies and adequate infrastructure, BEVs can meet daily mobility demands efficiently and sustainably, supporting broader EV adoption in urban environments with driving patterns similar to those observed in the studied cities.