A utility plant is one of key unit plants in large refinery and petrochemical plants that generates steam and electricity and supplies them to the main plants according to their utility demands. The dynamic market situation has led to the frequent changes of the demands for the steam and electricity. Consequently, the frequent demand changes have led to the frequent changes of the operation conditions of utility plant. This paper proposes a real-time optimization methodology that determines the optimal operation conditions that reduce the energy cost of the utility plant and satisfy the utility demands at the same time. We have developed a utility plant model in open form for its flexibility in selection of decision variables. The case studies have shown that the proposed methodology can be applied to the optimization of a utility plant that has a variety of demand patterns for multi-period operations and energy cost saving can be achieved by 4-8.9% compared with usual operation.