We consider the scheduling of deferrable and interruptible charging demand for plug-in electric or hybrid-electric vehicles (PEVs/PHEVs) in the smart grid over a finite horizon (e.g., 8pm-6am). The grid, acting as a centralized controller, decides when to charge which vehicle such that the total power consumption is maintained within a safety charging threshold while as many as consumers are satisfied by the deadline. Given that the charging profiles of PEVs/PHEVs are not constant and roughly have a (truncated) prescribed triangle shape, the grid has to take into account such burstiness/unevenness in the scheduling of their charging process. We develop an automated discrete-time scheduling algorithm, which dynamically tracks an unevenness measure for each consumer's charging profile and gives priority to consumers with highest unevenness that the grid can tolerate in each time slot. The unevenness measure is defined by the cumulative difference between the charging profile in each time slot and the average residual demand of his remaining charging profile. We compare this dynamic scheduling algorithm with (i) a similar algorithm using an unevenness measure defined by the average demand of each consumer during the entire finite horizon, and (ii) the SRPT algorithm giving priorities to consumers with shortest demand period and not taking into account unevenness. Simulations show that the dynamic algorithm can better avoid burstiness in the charging process and also satisfy many more consumers by the end of the finite horizon.