Ever increasing Plug-In Electric Vehicles (PEVs) can be considered as mobile storage units and can be exploited for grid ancillary services such as frequency support, voltage regulation and to support to intermittent renewable sources. However, available PEV power is limited by battery State-of-Charge (SoC) and customer flexibility. Smart charging control strategies are required in order to maximize PEV storage utilization. In this article, a new control strategy is developed to achieve flat load profile and to minimize the cost of charging. Both utility and customer benefits are given equal importance while utilizing PEV's storage for load flattening. Water Filling Algorithm (WFA) is used to distribute available PEV energy which in turn helps in the effective day-ahead scheduling of PEVs. Minimization of the cost of charging and maximization of PEV power usage has been accomplished by Adaptive Neuro-Fuzzy Inference System (ANFIS). ANFIS is trained to prioritize vehicles based on both utility and customer perspectives simultaneously. The impact of ANFIS prioritization on aggregate PEVs power availability and load flattening at a given time is studied. Also, the role of WFA in the prior scheduling of PEVs on load flattening is analyzed. Danish power grid data is used to implement the proposed control strategy in a residential distribution network. Due to the active power transactions between PEV and grid, the bus voltages are maintained within limits without using additional control for voltage regulation. © 2018 Elsevier B.V.