% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@PHDTHESIS{Thei:1017789,
author = {Theiß, Alina},
othercontributors = {Schneider, Michael David and Peis, Britta},
title = {{M}etaheuristic optimization for complex routing problems},
school = {Rheinisch-Westfälische Technische Hochschule Aachen},
type = {Dissertation},
address = {Aachen},
publisher = {RWTH Aachen University},
reportid = {RWTH-2025-07568},
pages = {1 Online-Ressource : Illustrationen},
year = {2025},
note = {Veröffentlicht auf dem Publikationsserver der RWTH Aachen
University; Dissertation, Rheinisch-Westfälische Technische
Hochschule Aachen, 2025},
abstract = {Optimizing transportation systems has become essential for
addressing today's logistics challenges. As global trade
grows consistently and consumer expectations for faster and
on-time deliveries rise, companies face increasing pressure
to deliver quickly and cost-efficiently. This thesis
addresses four routing problems: First, the single truck and
trailer routing problem with satellite depots (STTRPSD),
which can be used to model the problem of optimizing routes
of mail carriers in the last mile delivery stage of a mail
delivery network. Second, the vehicle routing problem with
depot operation constraints (VRP-DOC), which additionally
includes the assignment of households to mail carriers in
the route planning and incorporates depot operations. Third,
the angular-metric traveling salesman problem (ATSP) and the
angular-distance-metric traveling salesman problem (ADTSP),
relevant for minimizing sharp turns in the routing of heavy
vehicles. Last, the capacitated team orienteering problem
with time-dependent and piecewise-linear score functions
(C-TOP-TDPLSF) often used in the context of customer-focused
deliveries. To efficiently solve realistically sized
instances of these NP-hard problems, we use metaheuristics
like iterated local search or tabu search. Each heuristic is
designed to incorporate problem specific features to enhance
their performance. Extensive computational experiments show
significant improvements compared to state-of-the-art
algorithms from the literature and practices implemented in
the real world. For the STTRPSD, we use its natural
decomposition into subproblems in the design of our
heuristic, that reduces the travel times of real-world
solutions currently used in practice by our industry partner
on average by approximately $2\%.$ Addressing the VRP-DOC,
our work is one of the first ones to incorporate depot
operations into the route planning. In our algorithm, we use
problem-specific neighborhood operators and incorporate the
instance structure of real-world street networks. On
real-world instances, our heuristic is not only reducing
total travel times by approximately $6.5\%$ compared to the
currently implemented solutions from our industry partner
but also provides significantly simpler solutions with
regards to the letter handling operations at the depot,
highlighting the operational benefits of considering depot
operation constraints. For the ATSP and the ADTSP, we
incorporate the geometric features of the problems. Our
heuristic provides a good trade-off between runtime and
solution quality, and we find new best-known solutions for
around $80\%$ of benchmark instances for which an optimal
solution was not available. For the C-TOP-TDPLSF, our
heuristic is tailored to the specific structure of the score
function. Through in-depth analysis of the obtained
solutions, we provide practical recommendations to
companies, offering insights into improving operational
efficiency and decision-making.},
cin = {813210},
ddc = {330},
cid = {$I:(DE-82)813210_20140620$},
typ = {PUB:(DE-HGF)11},
doi = {10.18154/RWTH-2025-07568},
url = {https://publications.rwth-aachen.de/record/1017789},
}
Gast ::