We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
The green marine waste collector routing optimization with puma selection-based neighborhood search algorithm
AI summary Read the abstract
A mixed-integer optimization model combined with a novel hybrid Puma-neighborhood search algorithm was developed for routing marine debris collection fleets, reducing fuel and labor costs by 10–15% compared to conventional methods. GNOME trajectory software was used to estimate debris coordinates, accounting for the dynamic nature of floating waste.
Abstract Improper waste disposal by humans has created significant environmental issues in the marine ecosystem, including endangering aquatic life and accelerating the extinction of certain marine species. Due to the floating nature of the marine debris, the coordinates for collecting activities must be estimated in advance. In this article, GNOME software is used to estimate the coordinates of debris, and then a fleet of several ships is used to collect them. Also, a mixed integer linear programming model is presented for the routing optimization of debris collection fleets. The proposed optimization model formulates the objective function based on numerous factors, including labor cost, rent, and ship insurance, and considers constraints on fuel tank capacity, the time window, and the ship’s cargo capacity. A new hybrid algorithm combining the Puma algorithm and neighborhood search is proposed to address the problem. Metropolis acceptance is used in the simulated annealing algorithm to avoid the local optima and greedy selection. Numerical examples of the marine survey and the port of Rotterdam are used to test the proposed approach, which has been proven effective in several scenarios. Results achieved from the proposed hybrid method demonstrate considerable performance improvement in solving the problem. This approach has decreased total fuel and labor costs by 10–15% compared to conventional methods, with minimized time window violation reaching 25%. These results show a significant reduction in total operational costs with proper scheduling and route planning.
More Papers Like This
Two-Stage Optimization on Vessel Routing and Hybrid Energy Output for Marine Debris Collection
AI summary Read the abstract
Researchers propose a two-stage optimization approach for routing marine debris collection vessels equipped with hybrid solar-diesel energy systems, minimizing travel time and energy costs simultaneously. Results show hybrid energy vessels offer significant savings, and incorporating dynamic debris drift windows into routing models improves collection efficiency.
Plastic waste collection route optimization: Bulgan province
AI summary Read the abstract
Researchers applied travelling salesman problem and vehicle routing algorithms to optimize plastic waste collection in rural Bulgan province, Mongolia, identifying ten efficient routes across 19 locations and revealing that urban centers generate substantially more waste per kilometer than sparsely populated areas.
A Cross Disciplinary Framework for Cost-Benefit Optimization of Marine Litter Cleanup at Regional Scale
AI summary Read the abstract
This paper presents a cost-benefit framework for prioritizing where and how to clean up marine litter at regional scale. The framework considers factors like litter type, removal cost, ecological damage, and population exposure. Better economic tools for prioritizing cleanup could help direct limited resources toward areas where marine plastic removal has the greatest benefit.
Developing Strategy for Coastal Marine Debris Disposal Network Using Deep Reinforcement Learning
AI summary Read the abstract
Researchers used deep reinforcement learning to optimize the logistics network for collecting and disposing of coastal marine debris, with a focus on making the process economically viable without government subsidies. The model suggests that connecting debris collection to energy generation systems such as Styrofoam oil liquefaction can improve financial sustainability.
Microplastic sampling path planning for unmanned surface vehicle (USV) based on RRT * and Dijkstra fusion algorithm
AI summary Read the abstract
Researchers developed a path planning algorithm for unmanned surface vehicles (USVs) by fusing the RRT* and Dijkstra algorithms with B-spline path smoothing to enable autonomous multi-point microplastic sampling in coastal and surface waters, demonstrating superior performance over standalone RRT* and artificial potential field algorithms in terms of path length and smoothness.
Research digests by email
When a large batch of papers lands in the Atlas, we read through it and send a short write-up of what stood out.