This paper studies the electric vehicle (EV) charging scheduling problem where EV arrive at random unknown instants during the day with different charging demands and departure times. We consider single-phase charging EV in a three-phase charging station designed such that each EV has its own parking space. The objective is to build a real-time schedule that minimizes the total tardiness subject to the technical constraints of the charging station. We consider preemptive as well as non-preemptive EV charging. A mixed-integer linear programming (MILP) model is formulated for the offline problem. To solve the online problem, we propose heuristics based on the priority rule. Further, a local search is implemented to improve the objective value of the preemptive EV charging. Simulation results show that the proposed solving approaches outperform the existing heuristics developed in the literature. Moreover, we show that total tardiness is significantly reduced when preemption is exploited.