Solving large batches of traveling salesman problems with parallel and distributed computing

S. G. Ozden, A. E. Smith, K. R. Gue

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

In this paper, we describe and compare serial, parallel, and distributed solver implementations for large batches of Traveling Salesman Problems using the Lin–Kernighan Heuristic (LKH) and the Concorde exact TSP Solver. Parallel and distributed solver implementations are useful when many medium to large size TSP instances must be solved simultaneously. These implementations are found to be straightforward and highly efficient compared to serial implementations. Our results indicate that parallel computing using hyper-threading for solving 150- and 200-city TSPs can increase the overall utilization of computer resources up to 25% compared to single thread computing. The resulting speed-up/physical core ratios are as much as ten times better than a parallel and concurrent version of the LKH heuristic using SPC3 in the literature. For variable TSP sizes, a longest processing time first heuristic performs better than an equal distribution rule. We illustrate our approach with an application in the design of order picking warehouses.

Original languageEnglish (US)
Pages (from-to)87-96
Number of pages10
JournalComputers and Operations Research
Volume85
DOIs
StatePublished - Sep 1 2017

Fingerprint

Parallel and Distributed Computing
Traveling salesman problem
Warehouses
Distributed computer systems
Parallel processing systems
Travelling salesman problems
Batch
Heuristics
Processing
Order picking
Parallel Computing
Thread
Concurrent
Speedup
Resources
Parallel computing
Distributed computing
Computing

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Modeling and Simulation
  • Management Science and Operations Research

Cite this

@article{ad4bfc6d2d5c49e69d19bb4e74c5ad4b,
title = "Solving large batches of traveling salesman problems with parallel and distributed computing",
abstract = "In this paper, we describe and compare serial, parallel, and distributed solver implementations for large batches of Traveling Salesman Problems using the Lin–Kernighan Heuristic (LKH) and the Concorde exact TSP Solver. Parallel and distributed solver implementations are useful when many medium to large size TSP instances must be solved simultaneously. These implementations are found to be straightforward and highly efficient compared to serial implementations. Our results indicate that parallel computing using hyper-threading for solving 150- and 200-city TSPs can increase the overall utilization of computer resources up to 25{\%} compared to single thread computing. The resulting speed-up/physical core ratios are as much as ten times better than a parallel and concurrent version of the LKH heuristic using SPC3 in the literature. For variable TSP sizes, a longest processing time first heuristic performs better than an equal distribution rule. We illustrate our approach with an application in the design of order picking warehouses.",
author = "Ozden, {S. G.} and Smith, {A. E.} and Gue, {K. R.}",
year = "2017",
month = "9",
day = "1",
doi = "10.1016/j.cor.2017.04.001",
language = "English (US)",
volume = "85",
pages = "87--96",
journal = "Computers and Operations Research",
issn = "0305-0548",

}

Solving large batches of traveling salesman problems with parallel and distributed computing. / Ozden, S. G.; Smith, A. E.; Gue, K. R.

In: Computers and Operations Research, Vol. 85, 01.09.2017, p. 87-96.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Solving large batches of traveling salesman problems with parallel and distributed computing

AU - Ozden, S. G.

AU - Smith, A. E.

AU - Gue, K. R.

PY - 2017/9/1

Y1 - 2017/9/1

N2 - In this paper, we describe and compare serial, parallel, and distributed solver implementations for large batches of Traveling Salesman Problems using the Lin–Kernighan Heuristic (LKH) and the Concorde exact TSP Solver. Parallel and distributed solver implementations are useful when many medium to large size TSP instances must be solved simultaneously. These implementations are found to be straightforward and highly efficient compared to serial implementations. Our results indicate that parallel computing using hyper-threading for solving 150- and 200-city TSPs can increase the overall utilization of computer resources up to 25% compared to single thread computing. The resulting speed-up/physical core ratios are as much as ten times better than a parallel and concurrent version of the LKH heuristic using SPC3 in the literature. For variable TSP sizes, a longest processing time first heuristic performs better than an equal distribution rule. We illustrate our approach with an application in the design of order picking warehouses.

AB - In this paper, we describe and compare serial, parallel, and distributed solver implementations for large batches of Traveling Salesman Problems using the Lin–Kernighan Heuristic (LKH) and the Concorde exact TSP Solver. Parallel and distributed solver implementations are useful when many medium to large size TSP instances must be solved simultaneously. These implementations are found to be straightforward and highly efficient compared to serial implementations. Our results indicate that parallel computing using hyper-threading for solving 150- and 200-city TSPs can increase the overall utilization of computer resources up to 25% compared to single thread computing. The resulting speed-up/physical core ratios are as much as ten times better than a parallel and concurrent version of the LKH heuristic using SPC3 in the literature. For variable TSP sizes, a longest processing time first heuristic performs better than an equal distribution rule. We illustrate our approach with an application in the design of order picking warehouses.

UR - http://www.scopus.com/inward/record.url?scp=85017448781&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85017448781&partnerID=8YFLogxK

U2 - 10.1016/j.cor.2017.04.001

DO - 10.1016/j.cor.2017.04.001

M3 - Article

AN - SCOPUS:85017448781

VL - 85

SP - 87

EP - 96

JO - Computers and Operations Research

JF - Computers and Operations Research

SN - 0305-0548

ER -