Digitala Vetenskapliga Arkivet

Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Optimizing Transport Mission Planning with Large Language Models: A Comparative Analysis of the Traveling Salesman Problem
Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
2024 (English)Independent thesis Advanced level (degree of Master (One Year)), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

With the significant improvement in the task performance ability of Large Language Models (LLMs) through natural language processing, the possibility that computers can replace tasks previously thought to be performable only by humans is becoming increasingly realistic. This study seeks to evaluate the effectiveness of LLMs in solving transportation mission planning optimization problems, focusing on the Traveling Salesman Problem (TSP) as a typical example. The primary research question of this study is: “Can LLMs provide a feasible solution to the transport mission planning optimization problem?” To address this, we employ a controlled experiment as the main methodology using the Optimization by PROmpting (OPRO) (Yang et al., 2023) approach to tackle the Traveling Salesman Problem (TSP). We also explore the potential capabilities of large language models (LLMs) and investigate performance improvements by applying Chain-of-Thought (CoT) (Wei et al., 2022) prompting techniques. The CoT technique enables LLMs to systematically execute a series of logical steps in the problem-solving process, leading to accurate and efficient solutions for more complex optimization problems. As a result of the experiment, it is confirmed that performance is significantly improved through the CoT technique (Wei et al., 2022) as the situation becomes more complicated, with the optimality gap decreasing by 20 percentage points at 15 nodes. This highlights the contribution of the quality of task descriptions provided to LLMs. In conclusion, this study evaluated the optimization problem-solving ability of LLMs for the TSP based on the OPRO (Yang et al., 2023) and the CoT technique (Wei et al., 2022). This research presents a new approach for transport mission planning optimization, indicating the need for deeper analysis in follow-up studies to assess the scope of application and performance of LLMs. Follow-up research will further enhance LLMs’ optimization problemsolving ability and usability in actual transport mission planning.

Place, publisher, year, edition, pages
2024.
Keywords [en]
KeyLarge Language Model, Artificial Intelligence, Traveling Salesman Problem, Transport Mission Planning, Optimization, Prompt Engineering, Chainof- Thought
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:su:diva-242785OAI: oai:DiVA.org:su-242785DiVA, id: diva2:1955717
Available from: 2025-04-30 Created: 2025-04-30

Open Access in DiVA

fulltext(617 kB)15 downloads
File information
File name FULLTEXT01.pdfFile size 617 kBChecksum SHA-512
02a81cc8826fd417f91b34490e356880a05993aab4c933a9c998aae2c0a99e6a16cefb58548d31773dd9720ef8dc97ba839e984f3d6a8b31e8f50bb7f646bf17
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Kim, Michelle
By organisation
Department of Computer and Systems Sciences
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 15 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 26 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf