부산대학교 도서관

This paper addresses the problem of scheduling commodities in a multi-product pipeline. It is developed an auxiliary model for the operational decision-making process. It is considered a study upon a tank farm and a pipeline connecting a refinery to a harbour. The objective is the operating cost minimisation. This cost is determined by storage tank changeovers, by seguencing of activities, by plug product usage, and by electric pumping cost. It is employed the classical scheduling approach based on mixed integer linear programming with uniform time discretization. The proposed model is able to predict operating procedures based on a computer-optimised structure. The current pipeline operation is based on experience. Finally, simulation examples demonstrate a significant cost saving achieved by this scheduling methodology.