This paper presents a solution to Unit commitment (UC) of thermal units based on a new evolutionary algorithm known as Shuffled Frog Leaping Algorithm(SFLA). The integer coded algorithm is based on the behavior of group of frogs searching for a location that has the maximum amount of available food. This algorithm involves local search and shuffling process and these are repeated until a required convergence is reached. In this proposed method of SFLA for the UC problem, the scheduling variables are coded as integers, so that the minimum up/down time constraints can be handled directly. To verify the performance of the proposed algorithm it is applied to the 10 unit system for a one day scheduling period. © 2012 IEEE.