This paper proposes a simulation based methodology to assess plug-in hybrid vehicle (PHEV) behavior over 24-hour periods. Several representative 24-hour missions comprise naturalistic cycle data and information about vehicle resting time. The data were acquired during Filed Operational Tests (FOT) of a fleet of passenger vehicles carried out by the University of Michigan Transportation Research Institute (UMTRI) for safety research. Then, PHEV behavior is investigated using a simulation with two different charging scenarios: (1) Charging overnight; (2) Charging whenever possible. Charging/discharging patterns of the battery as well as trends of charge depleting (CD) and charge sustaining (CS) modes at each scenario were assessed. Series PHEV simulation is generated using Powertrain System Analysis Toolkit (PSAT) developed by Argonne National Laboratory (ANL) and in-house Matlab codes. Certification driving cycles, such as UDDS, HWFET, and US06, have been used in most previous research of PHEVs. Our results indicate very different instantaneous loads with naturalistic driving schedules. In particular, the required power of the motor frequently reached the maximum (120 kW), thus, the battery must be capable of supplying at least 134 kW power. In severe cases, the battery SOC dropped so fast to the lower limit (0.4) that the driving mode changed from the CD mode to the CS mode before the end of a one-way trip. Detailed time history of the battery SOC and current is predicted for the aggressive real-world driving conditions. This information provides valuable input for PHEV design optimization, battery health studies, assessment of real-world all electric range (AER) and charging schedules.