Plug-in electric vehicle (PEV) storage capacity is limited, and also, it depends on the battery energy level and trip timings. Hence, in order to effectively utilize PEV's storage capacity for grid support, smart charging and discharging control strategies are required. In this work, a new PEV control strategy is developed to achieve flat load profile and voltage regulation using PEV's storage capacity in an active residential distribution network. Both utility and PEV owner benefits (maximization of PEV usage and customer revenue) are given equal importance while scheduling PEVs for grid support. PEV prioritization is accomplished using an adaptive neuro-fuzzy inference system (ANFIS) with five decision variables. It has been assumed that the PEVs are available as per the scheduled timings, and vehicle prioritization may cause marginal shift in pre-scheduled times, but still target system on a chip is always ensured. During utilization of PEVs for load flattening, the voltage regulation at each bus, where PEVs are connected, is achieved by controlling active power transactions between the bus and PEVs. A multi-objective genetic algorithm is used to decide optimal power transaction between the grid and PEVs while maximizing PEV's storage exploitation without violating voltage limits. The proposed ANFIS prioritization (fixed rate) is compared with the variable-power-rate strategy in order to investigate the advantages of the proposed method. © 2010-2012 IEEE.