A Polynomial-Time Heuristic for the Travelling Salesman Problem Verified Against Held-Karp

Fuente: Zenodo
Salvato in:
Dettagli Bibliografici
Autore principale: Aggarwal, Minakshi
Natura: Recurso digital
Lingua:inglese
Pubblicazione: Zenodo 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866902163705298944
author Aggarwal, Minakshi
author_facet Aggarwal, Minakshi
contents <p><strong>This work presents a polynomial-time structural beam search algorithm for the Traveling Salesman Problem (TSP). The method combines bounded beam expansion with a restricted 2-opt repair step, ensuring polynomial bounds on both time and space. Experimental validation was carried out on symmetric, asymmetric, and blocked matrices for problem sizes up to n = 100. For n ≤ 15, the algorithm’s outputs were verified against the Held–Karp exact method, with complete agreement. Beyond this range, stress tests demonstrated empirical polynomial behavior, with runtime growth around O(n^{3.4}) and memory growth between O(n^{1.2}) and O(n^{1.7}). These results highlight the scalability and reliability of the approach, providing a new structural heuristic framework for combinatorial optimization</strong></p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_17018248
institution Zenodo
language eng
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle A Polynomial-Time Heuristic for the Travelling Salesman Problem Verified Against Held-Karp
Aggarwal, Minakshi
Travelling Salesperson Problem(TSP), Polynomial-Time Algorithms, Structural Beam Search, Combinatorial Optimization, Heuristics, Complexity Analysis
<p><strong>This work presents a polynomial-time structural beam search algorithm for the Traveling Salesman Problem (TSP). The method combines bounded beam expansion with a restricted 2-opt repair step, ensuring polynomial bounds on both time and space. Experimental validation was carried out on symmetric, asymmetric, and blocked matrices for problem sizes up to n = 100. For n ≤ 15, the algorithm’s outputs were verified against the Held–Karp exact method, with complete agreement. Beyond this range, stress tests demonstrated empirical polynomial behavior, with runtime growth around O(n^{3.4}) and memory growth between O(n^{1.2}) and O(n^{1.7}). These results highlight the scalability and reliability of the approach, providing a new structural heuristic framework for combinatorial optimization</strong></p>
title A Polynomial-Time Heuristic for the Travelling Salesman Problem Verified Against Held-Karp
topic Travelling Salesperson Problem(TSP), Polynomial-Time Algorithms, Structural Beam Search, Combinatorial Optimization, Heuristics, Complexity Analysis
url https://doi.org/10.5281/zenodo.17018248