A Two-Stage Decision-Support System for Floating Debris Collection in Reservoir Areas

The accumulation of floating debris is one of the main challenges of water conservancy projects, which may pollute the vulnerable ecosystem of the reservoir area and impose significant risks on waterway transportation and dam operations. Due to the dynamicity and uncertainty caused by water flow, the collection of floating debris is much more complicated than on-land waste collection. In this paper, we propose a two-stage decision-support system to optimize the task allocation and routing decisions for floating debris collection in the reservoir area, where the first stage is proactive planning based on historical/observed data and the second stage is reactive planning based on real-time data. The primary objective is to minimize the total collection cost while simultaneously ensuring the accumulation areas with high risks are prioritized in the daily collection plan, and both genetic algorithm and simulated annealing algorithm are used to solve the optimization problems. The proposed method is validated with a real-world case study at Wushan County in the Three Gorges Reservoir area. The computational results show that the level of time-dependent penalty cost on service priority, the types of collection ships, and the number and locations of unloading points are important influencing factors to the cost and responsiveness. Furthermore, the proposed two-stage decision-support system can help effectively optimize the operational planning of floating debris collection in reservoir areas.