This paper presents an application of an improved Genetic Algorithm for the unit commitment (UC) problem. The optimization is based on three different objective functions which are cost, reliability and an appropriate combination of them. This method is implemented on IEEE-RTS test system. Three different amount of required spinning reserve is used in optimization mode as a constraint. The results show that the proposed GA in C & R mode optimization is an effective method to solve the UC problem and also considering the reliability makes results more practical to use in restructured power system

Keywords: Unit commitment, Cost optimization, Reliability, Genetic algorithm