Marvin Kastner, M.Sc.

Address

Hamburg University of Technology
Institute of Maritime Logistics
Am Schwarzenberg-Campus 4 (D)
21073 Hamburg

 

Contact Details & Profiles

Office: building D room 5.007
Phone: +49 40 42878 4793
E-mail: marvin.kastner(at)tuhh(dot)de
ORCiD: 0000-0001-8289-2943
LinkedIn: https://www.linkedin.com/in/marvin-kastner/
ResearchGate: https://www.researchgate.net/profile/Marvin-Kastner
Google scholar: https://scholar.google.de/citations?user=lAR-oVAAAAAJ&hl=de&oi=ao
Scopus: https://www.scopus.com/authid/detail.uri?authorId=57221938031



Research Focus

  • Simulation-based Design of Container Terminals
  • Optimization of Yard Operations at Container Terminals
  • Data-driven Improvement of Maritime Security
  • Machine Learning in Maritime Logistic
  • Optimization of Multivariate Black-box Functions

Presentations and workshops (Excerpt)

  • 25.01.2023 a talk at the 7. Suderburger Logistics Forum: "AI-assisted planning of cargo handling facilities with the example of container terminals" (title translated)
  • 15.09.2022 a talk at the MLE-Days 2022: "Synthetic data for reinforcement learning in container terminal control systems."
  • 28.06.2022 a workshop at the Graduate Academy of TUHH: "Introduction to Jupyter Notebooks" (title translated) [more]
  • 02.07.2021 a workshop at the MLE-Days 2021: "Machine Learning in Maritime Logistics" (title translated) [zip]
  • 16.03.2021 a workshop at the Graduate Academy of TUHH: "Introduction to Jupyter Notebooks" (title translated) [more]
  • 30.11.2020 in the lecture series "Train Your Engineering Network" of the MLE initiative: "How to Talk About Machine Learning with Jupyter Notebooks"
  • 22.11.2019 at DISRUPT NOW! AI for Hamburg: "Artificial Intelligence in Maritime Economy" (title translated) [more]
  • 29.10.2019 in the context of forschungsbörse: "Maritime Logistics - an all-round cover" (title translated) [more]
  • 23.10.2019 at the Open Access Week 2019 at TUHH: "Data Analysis - Describe and Visualize Data with Jupyter Notebooks" (title translated) [more] [git]
  • 16.11.2018 at the GI DevCamp Hamburg: "Mobility Research and GDPR"
  • 27.09.2018 at SGKV WG regarding truck arrivals: "Forecasting and Neural Networks – What is possible?" (title translated)


Publications (Excerpt)

2024

[182410]
Title: Assessing performance of container slot allocation heuristics. <em>Adapting to the Future: Maritime and City Logistics in the Context of Digitalization and Sustainability</em>
Written by: Kastner, Marvin and Peters, Malte and Jahn, Carlos
in: (2021).
Volume: Number:
on pages: 427-454
Chapter:
Editor: In Ringle, Christian M. and Kersten, Wolfgang and Jahn, Carlos (Eds.)
Publisher:
Series:
Address:
Edition:
ISBN: 9783754927717
how published:
Organization:
School:
Institution:
Type:
DOI: 10.15480/882.4003
URL:
ARXIVID:
PMID:

[pdf]

Note:

Abstract: Purpose: In the last decades, the transport capacities of container vessels have tremendously increased. This leads to longer berth times and greater peak loads at container terminals, especially when schedules are perturbed. Thus, existing container handling processes need to be re-evaluated regarding their adequacy. Methodology: In the first step, the current literature is reviewed: which methods have been used for container slot allocation? In the second step, a simulation study is set up to compare two rule-based heuristics of Güven and Türsel Eliiyi (2014; 2019) with Levelling Stacking and Random Stacking. Findings: It is shown that the two rule-based heuristics of Güven and Türsel Eliiyi lead to shorter berth times than Levelling Stacking or Random Stacking. At the same time, the last two approaches show a clear superiority in workload balancing. The joint storage of container groups at Güven and Türsel Eliiyi leads to congestion at the stacking cranes in both cases for peak loads. Originality: This study is the first to compare these four stacking policies. For generating realistic container flows, data from an existing container terminal have been used. Previously unreported performance indicators are used for comparison. Thus, this study provides new insights for improved rule-based heuristics in future

2023

[182410]
Title: Assessing performance of container slot allocation heuristics. <em>Adapting to the Future: Maritime and City Logistics in the Context of Digitalization and Sustainability</em>
Written by: Kastner, Marvin and Peters, Malte and Jahn, Carlos
in: (2021).
Volume: Number:
on pages: 427-454
Chapter:
Editor: In Ringle, Christian M. and Kersten, Wolfgang and Jahn, Carlos (Eds.)
Publisher:
Series:
Address:
Edition:
ISBN: 9783754927717
how published:
Organization:
School:
Institution:
Type:
DOI: 10.15480/882.4003
URL:
ARXIVID:
PMID:

[pdf]

Note:

Abstract: Purpose: In the last decades, the transport capacities of container vessels have tremendously increased. This leads to longer berth times and greater peak loads at container terminals, especially when schedules are perturbed. Thus, existing container handling processes need to be re-evaluated regarding their adequacy. Methodology: In the first step, the current literature is reviewed: which methods have been used for container slot allocation? In the second step, a simulation study is set up to compare two rule-based heuristics of Güven and Türsel Eliiyi (2014; 2019) with Levelling Stacking and Random Stacking. Findings: It is shown that the two rule-based heuristics of Güven and Türsel Eliiyi lead to shorter berth times than Levelling Stacking or Random Stacking. At the same time, the last two approaches show a clear superiority in workload balancing. The joint storage of container groups at Güven and Türsel Eliiyi leads to congestion at the stacking cranes in both cases for peak loads. Originality: This study is the first to compare these four stacking policies. For generating realistic container flows, data from an existing container terminal have been used. Previously unreported performance indicators are used for comparison. Thus, this study provides new insights for improved rule-based heuristics in future

2022

[182410]
Title: Assessing performance of container slot allocation heuristics. <em>Adapting to the Future: Maritime and City Logistics in the Context of Digitalization and Sustainability</em>
Written by: Kastner, Marvin and Peters, Malte and Jahn, Carlos
in: (2021).
Volume: Number:
on pages: 427-454
Chapter:
Editor: In Ringle, Christian M. and Kersten, Wolfgang and Jahn, Carlos (Eds.)
Publisher:
Series:
Address:
Edition:
ISBN: 9783754927717
how published:
Organization:
School:
Institution:
Type:
DOI: 10.15480/882.4003
URL:
ARXIVID:
PMID:

[pdf]

Note:

Abstract: Purpose: In the last decades, the transport capacities of container vessels have tremendously increased. This leads to longer berth times and greater peak loads at container terminals, especially when schedules are perturbed. Thus, existing container handling processes need to be re-evaluated regarding their adequacy. Methodology: In the first step, the current literature is reviewed: which methods have been used for container slot allocation? In the second step, a simulation study is set up to compare two rule-based heuristics of Güven and Türsel Eliiyi (2014; 2019) with Levelling Stacking and Random Stacking. Findings: It is shown that the two rule-based heuristics of Güven and Türsel Eliiyi lead to shorter berth times than Levelling Stacking or Random Stacking. At the same time, the last two approaches show a clear superiority in workload balancing. The joint storage of container groups at Güven and Türsel Eliiyi leads to congestion at the stacking cranes in both cases for peak loads. Originality: This study is the first to compare these four stacking policies. For generating realistic container flows, data from an existing container terminal have been used. Previously unreported performance indicators are used for comparison. Thus, this study provides new insights for improved rule-based heuristics in future

2021

[182410]
Title: Assessing performance of container slot allocation heuristics. <em>Adapting to the Future: Maritime and City Logistics in the Context of Digitalization and Sustainability</em>
Written by: Kastner, Marvin and Peters, Malte and Jahn, Carlos
in: (2021).
Volume: Number:
on pages: 427-454
Chapter:
Editor: In Ringle, Christian M. and Kersten, Wolfgang and Jahn, Carlos (Eds.)
Publisher:
Series:
Address:
Edition:
ISBN: 9783754927717
how published:
Organization:
School:
Institution:
Type:
DOI: 10.15480/882.4003
URL:
ARXIVID:
PMID:

[pdf]

Note:

Abstract: Purpose: In the last decades, the transport capacities of container vessels have tremendously increased. This leads to longer berth times and greater peak loads at container terminals, especially when schedules are perturbed. Thus, existing container handling processes need to be re-evaluated regarding their adequacy. Methodology: In the first step, the current literature is reviewed: which methods have been used for container slot allocation? In the second step, a simulation study is set up to compare two rule-based heuristics of Güven and Türsel Eliiyi (2014; 2019) with Levelling Stacking and Random Stacking. Findings: It is shown that the two rule-based heuristics of Güven and Türsel Eliiyi lead to shorter berth times than Levelling Stacking or Random Stacking. At the same time, the last two approaches show a clear superiority in workload balancing. The joint storage of container groups at Güven and Türsel Eliiyi leads to congestion at the stacking cranes in both cases for peak loads. Originality: This study is the first to compare these four stacking policies. For generating realistic container flows, data from an existing container terminal have been used. Previously unreported performance indicators are used for comparison. Thus, this study provides new insights for improved rule-based heuristics in future

2020
[182410]
Title: Assessing performance of container slot allocation heuristics. <em>Adapting to the Future: Maritime and City Logistics in the Context of Digitalization and Sustainability</em>
Written by: Kastner, Marvin and Peters, Malte and Jahn, Carlos
in: (2021).
Volume: Number:
on pages: 427-454
Chapter:
Editor: In Ringle, Christian M. and Kersten, Wolfgang and Jahn, Carlos (Eds.)
Publisher:
Series:
Address:
Edition:
ISBN: 9783754927717
how published:
Organization:
School:
Institution:
Type:
DOI: 10.15480/882.4003
URL:
ARXIVID:
PMID:

[pdf]

Note:

Abstract: Purpose: In the last decades, the transport capacities of container vessels have tremendously increased. This leads to longer berth times and greater peak loads at container terminals, especially when schedules are perturbed. Thus, existing container handling processes need to be re-evaluated regarding their adequacy. Methodology: In the first step, the current literature is reviewed: which methods have been used for container slot allocation? In the second step, a simulation study is set up to compare two rule-based heuristics of Güven and Türsel Eliiyi (2014; 2019) with Levelling Stacking and Random Stacking. Findings: It is shown that the two rule-based heuristics of Güven and Türsel Eliiyi lead to shorter berth times than Levelling Stacking or Random Stacking. At the same time, the last two approaches show a clear superiority in workload balancing. The joint storage of container groups at Güven and Türsel Eliiyi leads to congestion at the stacking cranes in both cases for peak loads. Originality: This study is the first to compare these four stacking policies. For generating realistic container flows, data from an existing container terminal have been used. Previously unreported performance indicators are used for comparison. Thus, this study provides new insights for improved rule-based heuristics in future

2019

[182410]
Title: Assessing performance of container slot allocation heuristics. <em>Adapting to the Future: Maritime and City Logistics in the Context of Digitalization and Sustainability</em>
Written by: Kastner, Marvin and Peters, Malte and Jahn, Carlos
in: (2021).
Volume: Number:
on pages: 427-454
Chapter:
Editor: In Ringle, Christian M. and Kersten, Wolfgang and Jahn, Carlos (Eds.)
Publisher:
Series:
Address:
Edition:
ISBN: 9783754927717
how published:
Organization:
School:
Institution:
Type:
DOI: 10.15480/882.4003
URL:
ARXIVID:
PMID:

[pdf]

Note:

Abstract: Purpose: In the last decades, the transport capacities of container vessels have tremendously increased. This leads to longer berth times and greater peak loads at container terminals, especially when schedules are perturbed. Thus, existing container handling processes need to be re-evaluated regarding their adequacy. Methodology: In the first step, the current literature is reviewed: which methods have been used for container slot allocation? In the second step, a simulation study is set up to compare two rule-based heuristics of Güven and Türsel Eliiyi (2014; 2019) with Levelling Stacking and Random Stacking. Findings: It is shown that the two rule-based heuristics of Güven and Türsel Eliiyi lead to shorter berth times than Levelling Stacking or Random Stacking. At the same time, the last two approaches show a clear superiority in workload balancing. The joint storage of container groups at Güven and Türsel Eliiyi leads to congestion at the stacking cranes in both cases for peak loads. Originality: This study is the first to compare these four stacking policies. For generating realistic container flows, data from an existing container terminal have been used. Previously unreported performance indicators are used for comparison. Thus, this study provides new insights for improved rule-based heuristics in future