The increasing demand for eco-friendly transportation has resulted in the emergence of hybrid electric vehicles (HEVs). To assess the performance of HEVs, the original equipment manufacturers (OEM) have been focusing on real-world driving cycles in addition to fixed driving cycles that are defined in the regulations. Therefore, the development and validation phases with simulations or chassis dyno testing require representative driving cycle profiles that can emulate real-world driving conditions. As a result, automatic cycle generation using real-world driving patterns has become very popular in the literature. In this paper, a novel representative cycle generation method was investigated to obtain similar vehicle behavior as on-road measurements. The proposed method utilizes the data from GPS-equipped vehicles to identify the driving pattern with the parameters that impact fuel consumption and to create a synthetic cycle that represents the typical driving behavior of a given region. Two different methods, statistical-based and Markov chain-based cycle generation were compared with a reference cycle in terms of driving pattern characteristics. After testing the most representative generated cycle on a chassis dynamometer, a 0.63% deviation was obtained in terms of fuel consumption compared to the reference cycle.