This paper presents a low-cost, scalable method that can schedule in real time the charging process of a large set of controllable plug-in electric vehicles within a finite spatial area such as a parking garage or residential neighborhood. This method allows the charging system to abide by the constraints on total power consumption, which can be either sent down from the upper-level utility operator or generated locally in response to electricity price variations and local measurements, while satisfying all the charging requirements from the customers. The charging system can collect information about the grid-connected vehicles and send charging commands (including on/off signals and charging rates) to individual chargers through an advanced metering infrastructure (AMI) or a local SCADA (supervisory control and data acquisition) system. The system can generate charging commands in such a way that the total power consumption by the collection of vehicles does not exceed the specified limitation and that the charging deadlines of all vehicles are met as well. The charging sequence is generated using a prioritization scheme based on several factors such as the status of vehicles, batteries and customers. The system can be part of a distribution management system and coordinate with an existing demand response management system.