Main content area

A simulation-based multi-agent particle swarm optimization approach for supporting dynamic decision making in marine oil spill responses

Ye, Xudong, Chen, Bing, Li, Pu, Jing, Liang, Zeng, Ganning
Ocean & coastal management 2019 v.172 pp. 128-136
algorithms, case studies, coastal zone management, decision making, energy use and consumption, models, oil spills, oils, process control, socioeconomic factors, system optimization, weathering, Atlantic Ocean
How to improve marine oil spill response efficiency to minimize environmental and socioeconomic impacts has been recognized as a growing, critical need worldwide in both scientific and practical fields. The efficiency much depends on how sound the response decisions can be made simultaneously at both systematic (or holistic) (e.g., best use of resources for the entire response system) and individual levels (e.g., optimal operation of skimmers on a spill site). This study proposed a new simulation-based multi-agent particle swarm optimization (SA-PSO) approach for supporting marine spill decision-making through the integrated simulation and optimization of response device allocation and process control. Agent-based modeling as an emerging simulation method was first applied for simulating oil spill fate and response. Particle swarm optimization method was further adopted to optimize response device/vessel allocation and performance with a minimal cost and time. Multi-agent system finally controlled and transmitted the results from agent-based modeling and particle swarm optimization as a dynamic and interactive system. The proposed method was tested by a hypothetical case study in the North Atlantic Ocean with consideration of oil weathering and non-weathering scenarios based on simplified conditions. Through the developed approach, the response time was reduced by 11.7% and 5.9% respectively under the two scenarios for vessel allocation and recovery operations with about 90% decrement of fuel consumption. The results showed the strong capability of the approach for decision makings in oil spill responses by recommending optimal management of resources and efficient response operation in a dynamic manner.