Please use this identifier to cite or link to this item: http://hdl.handle.net/1893/29359
Appears in Collections:Computing Science and Mathematics Conference Papers and Proceedings
Author(s): McMenemy, Paul
Veerapen, Nadarajen
Adair, Jason
Ochoa, Gabriela
Contact Email: paul.mcmenemy@stir.ac.uk
Title: Rigorous Performance Analysis of State-of-the-Art TSP Heuristic Solvers
Editor(s): Liefooghe, A
Paquete, L
Citation: McMenemy P, Veerapen N, Adair J & Ochoa G (2019) Rigorous Performance Analysis of State-of-the-Art TSP Heuristic Solvers. In: Liefooghe A & Paquete L (eds.) Evolutionary Computation in Combinatorial Optimization. Lecture Notes in Computer Science, 11452. EVOCOP 2019: European Conference on Evolutionary Computation in Combinatorial Optimization, Leipzig, Germany, 24.04.2019-26.04.2019. Cham, Switzerland: Springer International Publishing, pp. 99-114. https://doi.org/10.1007/978-3-030-16711-0_7
Issue Date: 2019
Series/Report no.: Lecture Notes in Computer Science, 11452
Conference Name: EVOCOP 2019: European Conference on Evolutionary Computation in Combinatorial Optimization
Conference Dates: 2019-04-24 - 2019-04-26
Conference Location: Leipzig, Germany
Abstract: Understanding why some problems are better solved by one algorithm rather than another is still an open problem, and the symmetric Travelling Salesperson Problem (TSP) is no exception. We apply three state-of-the-art heuristic solvers to a large set of TSP instances of varying structure and size, identifying which heuristics solve specific instances to optimality faster than others. The first two solvers considered are variants of the multi-trial Helsgaun's Lin-Kernighan Heuristic (a form of iterated local search), with each utilising a different form of Partition Crossover; the third solver is a genetic algorithm (GA) using Edge Assembly Crossover. Our results show that the GA with Edge Assembly Crossover is the best solver, shown to significantly outperform the other algorithms in 73% of the instances analysed. A comprehensive set of features for all instances is also extracted, and decision trees are used to identify main features which could best inform algorithm selection. The most prominent features identified a high proportion of instances where the GA with Edge Assembly Crossover performed significantly better when solving to optimality.
Status: AM - Accepted Manuscript
Rights: This is a post-peer-review, pre-copyedit version of a paper published in Liefooghe A & Paquete L (eds.) Evolutionary Computation in Combinatorial Optimization. Lecture Notes in Computer Science, 11452. EVOCOP 2019: European Conference on Evolutionary Computation in Combinatorial Optimization, Leipzig, Germany, 24.04.2019-26.04.2019. Cham, Switzerland: Springer International Publishing, pp. 99-114. The final authenticated version is available online at: https://doi.org/10.1007/978-3-030-16711-0_7

Files in This Item:
File Description SizeFormat 
McMenemy.pdfFulltext - Accepted Version3.73 MBAdobe PDFView/Open



This item is protected by original copyright



Items in the Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

If you believe that any material held in STORRE infringes copyright, please contact library@stir.ac.uk providing details and we will remove the Work from public display in STORRE and investigate your claim.