Combined cycle units represent the majority of new generating unit installations across the globe. Combined cycle units feature distinct advantages for power generation such as high efficiency, fast response, shorter installation time, abundance of gas, and environmental friendliness. The short-term scheduling of combined cycle units could represent a complicated optimization problem because combined cycle units could have multiple operating configurations based on the number and the status of combustion turbine and steam turbine. In this paper, we present a method for establishing the state space diagram of combined cycle units for applying dynamic programming and Lagrangian relaxation to the security constrained short-term scheduling problem. Although the proposed method is applied in this paper to analyze combined cycle units, it is suitable for other types of generating units with multiple configurations including fuel switching/blending units, constant/variable pressure units, and dual boiler units. The paper presents a few case studies to verify the advantages of combined cycle units in competitive electricity markets.