On the scheduling of real world multiproduct pipelines with simultaneous delivery

Abstract

Common carrier pipelines are the best suited mode for transporting large volumes of petroleum products from production areas to long-distance terminals. In such pipelines, batches of refined petroleum products are pumped back-to-back, without any physical barrier separating them. The problem addressed in this paper deals with the scheduling of a real world multi-product, single source pipeline system that must supply the products to several offtake terminals. Most contributions on the scheduling of single-source multiproduct pipeline operations deal with the sequence deliveries in distribution terminals i.e., at any time only one terminal is connected to the pipeline. Practically, pipeline operators usually carry out simultaneous deliveries to multiple terminals to cut down the number of pipeline stoppages and pump switchings so as to reduce the energy consumed for resuming flow in idle pipeline segments, and the pump maintenance costs. This paper introduces a novel continuous mathematical approach, mixed integer linear programming to solve the short-term operational planning by allowing the execution of simultaneous deliveries to multiple receiving terminals during a pumping operation. Contrarily to previous continuous approaches that perform the pipeline input/output operations in two hierarchical stages, the new formulation aims to find both of them in a single step. The objective of this work is to find the optimal sequence of input and output operations that satisfy terminal requirements at minimum total costs. As compared to previous works, significantly improvement in solution quality has been achieved. Especially, the proposed formulation leads to better utilization of the pipeline capacity, and consequently, a substantial reduction in the amount of required time for satisfying all of the specified product deliveries, i.e., the pipeline running time. The main results are presented.