Options
Deep learning assisted heuristics and exact methods for the vehicle routing problem with side constraints
Citation Link: https://doi.org/10.15480/882.15061
Publikationstyp
Doctoral Thesis
Date Issued
2025
Sprache
English
Author(s)
Advisor
Referee
Title Granting Institution
Technische Universität Hamburg
Place of Title Granting Institution
Hamburg
Examination Date
2025-03-26
Institute
TORE-DOI
Citation
Technische Universität Hamburg (2025)
This thesis addresses vehicle routing problems (VRP) with complex side constraints, focusing on developing deep learning-assisted heuristic methods that deliver near-optimal solutions. We propose novel approaches for the capacitated VRP with time windows by integrating graph convolutional neural networks into heuristic methods. These networks predict promising edges that are utilized to enhance the efficiency of heuristics. Additionally,we explore the integration of quantum-inspired computing within heuristic frameworks, designing a hybrid heuristic that combines deep learning with specialized quantum-inspired hardware to enhance scalability and solve larger instances more effectively.
Subjects
Vehicle Routin | Deep Learning | Heuristics | Neural Networks | Quantum-inspired Computing | Combinatorial Optimization
DDC Class
003: Systems Theory
004: Computer Sciences
005.1: Programming
620: Engineering
Publication version
publishedVersion
Loading...
Name
Dornemann_Jorin_Dissertation.pdf
Size
3.23 MB
Format
Adobe PDF