Optimization of Downstream Oil Supply Chain with Respect to Transport Cost Using Monte Carlo Simulation to Evaluate Disruption Scenarios

Abstract

This research explores the effects of random and anticipated disruptions on the transport costs associated with the volume flows between different stages of a downstream oil supply chain. How do the volume flows change when a refined oil supply chain’s operations get disturbed due to disruptions of various nature? The purpose of the study was to develop a Multi Integer Linear Programming model that would simulate a multi-echelon, multi-modal, and multiproduct refined oil supply chain. The researchers reviewed literature pertaining to the design of the supply chain to understand the disruptions that can create operational difficulties. Designing the model includes the refining and import facilities, storage depots and customer demand nodes while disruptions are used to influence the product flows between these entities. The study is divided into two parts, with a deterministic model and a Monte Carlo simulation. The deterministic model is used to achieve a baseline of the supply chain performance without any disruptions. Monte Carlo sampling is used to generate scenarios with disruptions in refined oil supply, increased demand, increase in transport costs, and interruption in pipeline services. The same model is used to simulate an optimized flow of products under all generated scenarios. The results showed an overall increase in the transport cost of the supply chain with drastic changes in flows between entities. There was an increase in the import of refined oil products to make up for the local production shortage, along with an increase in the use of bulk cargo modes to reduce costs wherever possible. Particularly road transport was more used to cover up the pipeline flows. The implication of this study is that random and anticipated disruptions in the supply chain greatly increase overall transport costs. More refined fuel has to be imported to fulfil the high demand and low local production. Other modes have to be utilized in areas where pipelines are not operational. The limitations of this study are that the supply chain is restricted to just one company and only considers the transport cost and disruptions in limited capacities are considered. Future research should examine the need for inventory management and alternate pipeline route development so that the robustness of a supply chain can be measured under disruption scenarios.