Optimal network design and storage management in petroleum distribution network under uncertainty

This paper studies the status of storage tanks in petroleum distribution network under uncertainty. It is difficult to describe the topology of this network precisely. As an instance, the capacity of pipelines and tanks due to corrosion cannot be presented as real numbers. Also production of oil fields, demand of costumers and food of refineries in a long-term programming are given under uncertainty. The contribution of this paper is to capture with the corresponding granular information applying fuzzy concepts. For this mission, an off-line basic framework is proposed in order to find the least daily-transportation-cost of crude oil flows through a capacitated network. To transform this management problem into a traditional minimal cost flow problem (MCFP), a T -floors network is provided which simulates the oil transportation in T days. The dummy links adjoining two consecutive floors are embedded as storage depots and their flows are corresponded to maintained flow. To exhibit with imprecision concerning this dynamic network, a fuzzifier can be applied which obtains a fuzzy number reflecting system description. Then a fully fuzzified MCFP is solved applying a ranking function understood by authors in previous works. We illustrate the performance of the proposed scheme on a simplified pilot of Iranian petroleum industry network. Furthermore, we offer a modified version of successive shortest path algorithm which is able to find the optimal place and the optimal capacity of storage tanks when they are required through the long-term programming period.