(680a) Maximum Resiliency Identification for the General Refinery Supply Chain

A general refinery supply chain (GRSC) can be divided into three subsystems. Specifically, the first subsystem involves the front-end crude movement operations, including unloading various crudes to storage tanks, crude pipeline transportation, crude blending at charging tanks, and feeding to crude distillation units (CDUs); the second subsystem is about the refinery manufacturing, which involves many unit operations such as fractionation, reaction, mixing, and storage management in the refinery; and the third subsystem is about the multi oil-product distribution, which consists of inventory management at the refinery product tank farm, multi oil-product pipeline transportation, as well as oil product receiving, exporting, and inventory management at different oil depots. GRSC is critically important for the profitability and sustainability of oil and gas companies. However, the GRSC is also susceptible to risks from various disruptions, especially threats of multi-natural disasters such as hurricanes, flooding, or tornado outbreaks, which bear low likelihood but have significant economic, environmental, and/or public health consequences. Obviously, disruptions from these multi-natural disasters shall be well adopted by the entire GRSC system; otherwise, they may cause not only economic losses to the industry but also threats or disasters to local environments and/or public health.

In this paper, a resilience characterized optimization model for a full-scale GRSC system has been developed. The optimization model covers operations of crude purchasing, shipping, unloading, storage, pipeline transportation, blending and discharging, refinery manufacturing, as well as oil product storage and distribution. Based on the optimization model, three quantitative resilience indexes have also been developed to characterize the maximum resilience capability of a GRSC system: (i) a quantity-based resilience index (QRI_n), (ii) a property-based resilience index (PRI_n), and (iii) the minimum selection of QRI_n and PRI_n, i.e., the combined residence index (CRI_n). From different viewpoints, the QRI_n, PRI_n, and CRI_n provide their maximum time durations during which a GRSC can at least sustain its minimum safe load operation under disruption impacts caused by multi-natural disasters or accidents. The three resilience indexes have been alternatively employed in the objective function of the GRSC optimization model for optimal solution identification. Their solution results provide: (1) the maximum potential of the GRSC to sustain its maximum run length; and (2) the information on how to simultaneously coordinate and control schedules and operations among the three subsystems. The development will help oil and gas companies to not only quantitively recognize the inherent resilience of their supply chains; but also have better decision support to deal with threats of multi-natural disasters and unpredictable accidents in proactive and cost-effective ways.