Joint modelling of electric vehicle charging and daily activity scheduling

Abstract

The use of Electric Vehicles (EVs) introduces new constraints and choice dimensions in daily mobility behaviour, requiring EV drivers to adapt their activity participation, start times, durations, and locations to accommodate charging needs. Existing approaches to evaluating EV charging policies and infrastructure planning often lack behavioural realism. By assuming homogeneous charging behaviour and ignoring its connection to activity-travel patterns, these approaches fail to capture nuanced responses to charging management policies (e.g., pricing) and infrastructure constraints. We introduce a new modelling approach that explicitly captures heterogeneous behaviours, effectively modelling the trade-offs between multiple-choice dimensions and policy responses. Unlike sequential decision models, our simultaneous model structure jointly addresses activity and charging schedule, characterised by activity and charge times (start time and duration), activity sequences, locations and charging modes (e.g. slow, fast, and super-rapid). By operationalising this simultaneous modelling framework, we analyse individual activity and charging behaviours, leveraging activity-based charging flexibility (e.g., home, work, and public locations) and responses to policies. Our findings demonstrate the variability in activity and charging behaviours across different use cases, highlighting the importance of considering diverse charging access and activity patterns.