Unit commitment (UC) problem is one of the medium term decision problems in power system and it is used to determine the mix/ combination of generators to be operated or to be idle. By solving the UC problem, once the binary data is obtained the next level is to find the optimum level of dispatch of each generators in the mix. For a large generating system UC program is a large scale, non linear mixed integer programming (MIP). This paper proposes a hybrid algorithm to solve UC problem by considering equality and inequality constraints. The objective of the algorithm is to minimize the fuel cost after meeting all the possible constraints. The proposed algorithm initially solved the UC problem by modified LR method. LR method most often gives sub optimal solution, so once the problem is solved by LR and if some of the output is infeasible then by solving it by heuristic technique such as priority list method, we can modify LR. The generation output level is determined by particle swarm optimization method. The algorithm was tested in four unit eight hour systems and ten unit twenty four hour system. The simulation output shows that the developed algorithm performed well and also applicable for large systems. © 2012 Praise Worthy Prize S.r.l. - All rights reserved.