With growing of the great interest in the energy problems, real-time pricing (RTP) for power systems has attracted attentions in the world. In our research, the new distributed optimization method is proposed for RTP. It is based on information exchanges between players (consumers, suppliers and distributors) through networks. In the past our research, the time required for information exchange is not considered. However, in actual systems, the time required for the wireless communication causes the time delay and instabilities of the system. Then this paper proposed to compensate for time-delay by prediction of demand-supply quantities.