| The solid waste that enters the ocean is called marine debris.It mainly comes from human activities.Because of its large quantity,wide distribution and great harm,which are the characteristics of marine debris,it is urgent to alleviate the pollution of marine debris.And countries around the world have spend a lot of money on recycling and cleaning up marine debris every year.In order to save energy and reduce emissions,it has become an important research project to study vessel routing optimization for marine debris collection.In the past,the routing optimization of waste collection is usually limited to the land,and there are few articles about the optimization of vessel routing for marine debris collection.In the ocean,due to the uninterrupted flow of sea water and the uninterrupted change of wind field,the location of marine debris is constantly changing.Through the analysis of domestic and foreign literatures,this paper finds that the initial location of marine debris can be retrieved through satellite monitoring images.The oil spill software(GNOME)combined with the marine dynamic model is used to simulate the movement trajectory of floating marine debris.When the input data is accurate,the scheme has high feasibility and reliability.In this paper,the flow field data from the Hybrid Coordinated Ocean Model(HYCOM),and the wind field data from the National Center for Environmental Prediction Global Forecast System(NCEP GFS)and related hydrological factors,such as temperature,salinity,seawater diffusion coefficient,etc.,were used to simulate and construct the marine environment,what combined with GNOME’s diagnostic model,forecast the position of marine debris within the time window.Considering the complexity of the actual situation,it is impossible to accurately know the weight of marine debris by observation or prediction of the location of marine debris.In this paper,the weight of marine debris is introduced into the model as a random variable,and a random opportunity constraint planning model is established,and the logistics network is used to collect marine debris.(1)A random opportunity constraint planning model with the minimum cost of the marine debris collection is established,and the carbon tax cost is considered in the model.In the model,the opportunity constraint of debris weight uncertainty,time window constraints and fuel capacity constraints of vessels is considered.A particle swarm optimization combined with adaptive large-scale neighborhood search is designed to solve this model.The debris dispersion located in the East China Sea near Shanghai Port is used as a case study to validate the proposed model.The result shows that the total cost of marine debris collection and the average distance between marine debris and the port are highly positively correlated.And there is a highly positive correlation between the carbon tax cost and average distance data.Compared with the worst-case scenario,the total cost of collecting marine debris by choosing the best time can be saved by 6 614.15 RMB and 6.17%,and the cost of carbon tax is saved by 62.84 RMB and8.11%.The different distribution of the weight of marine debris did not affect the results.Sensitivity analysis shows that the target is also insensitive to the carbon tax price.Under the same conditions,the calculated result of PSO+ALNS was 26.05% less than that of LNS.(2)A two-objective opportunity constrained model with the objective of minimizing the total time of marine debris collection and carbon emissions and the uncertain weight of marine debris is established.In the model,vessels speed variable,the opportunity constraint of debris weight uncertainty,time window constraints,fuel capacity constraints of vessels and total cost constraints is considered.A particle swarm optimization combined with adaptive large-scale neighborhood search and fast non-dominated sorting is designed to solve the model.The debris dispersion located in the East China Sea near Shanghai Port is used as a case study to validate the proposed model.The result shows that the correlation coefficient between the total time of marine debris collection and the average distance between marine debris and port is 0.62.The correlation coefficient is greater than 0.5 and less than 0.8,and the total time and the average distance are moderately correlated.The correlation coefficient between carbon emissions and the average distance data is 0.96.They are highly positive correlation.The correlation coefficient between the total time and carbon emissions is 0.80.They are highly positive correlation.The correlation between carbon emissions and average distance data is the highest.Taking the second day and the fourth day as an example,compared with the worst-case scenario,the total time of collecting marine debris by choosing the best time can be saved by 2.40 hours and3.59%,and carbon emissions is saved by 6.10 tons and 13.39%.The different distribution of the weight of marine debris did not affect the total time and carbon emissions.Sensitivity analysis shows that the use of vessels with a deadweight of 40 tons for marine debris collection can reduce the total cost on the one hand,improve the utilization rate of vessel deadweight on the other hand,and reduce the total time and carbon emissions of marine debris collection.Under the same conditions,BOPSO+ALNS can save 25.63% of the average total time of marine debris collection and 0.32% of the average total carbon emissions compared with BOLNS. |
