A hybrid algorithm on the vessel routing optimization for marine debris collection. (15th November 2021)

• Record Type: Journal Article
• Authors: Duan, Gang; Fan, Tao; Chen, Xiaohui; Chen, Li; Ma, Junfeng
• Is Part Of: Expert systems with applications. Volume 182(2021)
• Publication Date: 2021-11-15
• Subjects: Marine debris collection -- Vessel routing optimization -- GNOME -- Adaptive large-scale neighborhood search -- Wolf pack algorithm; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2021.115198
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: The logistics network is used to optimize the marine debris collection problem. GNOME software is employed to predict the trajectory of marine debris. The fuel cost in the objective is a piecewise linear function of the total weight of vessel load. A hybrid algorithm of adaptive large-scale neighborhood search and wolf pack algorithm is proposed. Some valuable managerial insights are drawn. Abstract: More and more marine debris not only damages the environment, but also endangers human health and marine life. Since the location of marine debris is not fixed, we first apply the GNOME software to predict the trajectory of marine debris and determine its location. Then, the logistics network is used to optimize the collection of marine debris. A mixed integer linear programming model for debris vessel routing is proposed. The objective function is to minimize the total cost, which includes the fuel cost consisting of is a piecewise linear function, labor cost, and fixed cost consisting of rent and insurance relative to the vessels used, taking into account the vessel weight capacity, fuel tank capacity, and time windows constraint at each debris location. We propose a hybrid algorithm of adaptive large-scale neighborhood search, combined with a wolf pack algorithm. This paper takes Yangtze River Estuary to the East China Sea as an example to verify the proposed model and algorithm. The results show that, compared with the worst-case, the most appropriate collection time can save up to 6.38% of the cost, which is about $75, 590 per calendar year. The sensitivity analysis of debris weight and vessel capacity is also performed.