Optimizing Oil Spill Emergency Logistics: A Time-Varying Multi-Resource Collaborative Scheduling Model

Abstract

Emergency resource scheduling is at the heart of the response to an oil spill, as it lays the foundation for all other emergency operations. Extant studies address the dynamicity inherent to these operations primarily by modeling a dynamic network flow with static data, which is not applicable to continuously changing conditions resulting from oil film movement. In order to enhance the responsiveness and cost-efficiency of the response to oil spills, this paper takes a novel approach and formulates a multi-objective location-routing model for multi-resource collaborative scheduling, namely, harnessing time-varying parameters rather than static data to model real-time changes in the demand for emergency resources and the transportation network. Additionally, the model considers various operational factors, including the transportation of multiple resources in the order of operating procedures; the coordination of split delivery with the consumption of emergency resources; and the matching of multiple resources with suitable vehicles. To solve the proposed model, a hybrid heuristic algorithm of PSO-PGSA is developed, which utilizes particle swarm optimization (PSO) to search widely for non-dominated solutions. The algorithm then makes use of the plant growth simulation algorithm (PGSA) to find the more effective vehicle routes based on the obtained solutions. Finally, a numerical analysis is used to illustrate the practical capabilities of the developed model and solution strategies. Most significantly, our work not only to validated the methodology proposed here, but also underlines the importance of incorporating the features of an oil spill emergency response into emergency logistics in general.

Supply chain network optimization for emergency materials considering demand satisfaction

“…How to make a scientific and accurate prediction of the required emergency materials under public emergencies is the first key link of emergency response (Lin et al, 2023), which is mainly categorized into two types of methods: qualitative and quantitative. Qualitative prediction is mainly based on Delphi's method, subjective probability method, combining experts' professional knowledge and subjective experience to estimate the result; common quantitative prediction methods mainly include case-based reasoning method (Platon et al, 2015;Zhang et al, 2022), neural network model (Lin et al, 2022;Yinghui et al, 2022;Zhang & Yang, 2015), gray system (Bo et al, 2015;Geng & Hou, 2021;Zhang et al, 2021;Zhongjun et al, 2019), and so forth, which quantify and predict the research object based on historical data. Focusing on public health emergencies, LIU et al coupled the dynamic demand prediction mechanism of medical resources based on the epidemic spread model with the optimal allocation and transportation mechanism of medical resources based on the multi-stage planning model and put forward a discrete spatio-temporal network model (Liu et al, 2015).…”