A hybrid genetic algorithm (HGA) is presented to solve the thermal unit commitment (UC). It effectively uses the binary-coded genetic algorithm incorporating a priority list ordering scheme to solve the unit commitment up/down problem and the lambda-iteration method to solve the power economic dispatch problem, This modification helps to explore the search space very effectively, It generates better solutions than the other methods. Problem formulation, representation and the simulation results with systems of up to 10 units and 24-h scheduling horizon are presented.