Understanding and predicting charging behaviour of electric vehicles' users is essential for the appropriate design of charging services and for the implementation aggregator services mediating between electric vehicle drivers, electricity markets, distribution system operator, and transmission system operator. Research into actionable charging behaviour insights to be implemented in predictive models has so far been modest. The present paper intends to contribute at the development of predictive models of charging patterns for operational implementation. It proposes hazard-based analysis of the gap times between charging events model with time dependent covariates in order to investigate which set of potential covariates that will enable an eventual formulation of short-term predictive model of the timing of charging events. Empirical estimation of the hazard model shows that both monitored vehicle state variables (e.g. state of charge, cumulative average driving speed) and individual characteristics significantly affect the instantaneous rate of occurrence of charging events.