dc.contributor.advisor | Kampffmeyer, Michael | |
dc.contributor.advisor | Bordin, Chiara | |
dc.contributor.author | Breimo, Marit Utheim | |
dc.date.accessioned | 2023-08-21T06:43:28Z | |
dc.date.available | 2023-08-21T06:43:28Z | |
dc.date.issued | 2023-06-04 | en |
dc.description.abstract | The transport sector is a major contributor to the emission of greenhouse gases and air pollution. As urbanization and population growth continue to increase, the demand for transportation services grows, emphasizing the need for sustainable practices. Therefore, incorporating sustainability into the transport sector can effectively reduce its negative impacts on the environment and optimize the utilization of resources.
This thesis aims to address this issue by proposing a novel method that integrates neural networks into the development of a green vehicle routing model. By incorporating environmental considerations, particularly fuel consumption, into the optimization process, the model seeks to generate more sustainable route solutions. The integration of machine learning techniques, specifically an attention-based neural network, demonstrates the potential of combining machine learning with operations research for effective route optimization.
While the effectiveness of the green vehicle routing problem (GVRP) has been demonstrated in providing sustainable routes, its practical applications in real-world scenarios are still limited. Therefore, this thesis proposes the implementation of the GVRP model in a real-world waste collection routing problem. The study utilizes data obtained from Remiks, a waste management company responsible for waste collection and handling in Tromsø and Karlsøy.
The findings of this study highlight the promising synergy between machine learning and operations research for further advancements and real-world applications. Specifically, the application of the GVRP approach to waste management issues has been shown to reduce emissions during the waste collection process compared to routes optimized solely for distance minimization. The attention-based neural network approach successfully generates routes that minimize fuel consumption, outperforming distance-optimized routes. These results underscore the importance of leveraging the GVRP to address environmental challenges while enhancing decision-making efficiency and effectiveness. Overall, this thesis provides insights for developing sustainable and optimized routes for real-world problems. | en_US |
dc.description | 23.08.23: Trekkes tilbake fra visning som løsning på at oppgaven ble ferdigstilt fra studieadministrasjonen litt for fort/IHTI | |
dc.identifier.uri | https://hdl.handle.net/10037/30103 | |
dc.language.iso | eng | en_US |
dc.publisher | UiT The Arctic University of Norway | en |
dc.publisher | UiT Norges arktiske universitet | no |
dc.rights.holder | Copyright 2023 The Author(s) | |
dc.rights.uri | https://creativecommons.org/licenses/by-nc-sa/4.0 | en_US |
dc.rights | Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) | en_US |
dc.subject.courseID | EOM-3901 | |
dc.subject | Environmental engineering | en_US |
dc.subject | Mathematics. Information and communication science. Physics. Optimization. Machine Learning. | en_US |
dc.title | Attention-Based Neural Network for Solving the Green Vehicle Routing Problem in Waste Management | en_US |
dc.type | Master thesis | en |
dc.type | Mastergradsoppgave | no |