DOI QR코드

DOI QR Code

외판원 문제의 지역 분할-연결 기법

Travelling Salesman Problem Based on Area Division and Connection Method

  • 이상운 (강릉원주대학교 과학기술대학 멀티미디어공학과)
  • Lee, Sang-Un (Dept. of Multimedia Eng., Gangneung-Wonju National University)
  • 투고 : 2014.10.28
  • 심사 : 2015.06.12
  • 발행 : 2015.06.30

초록

본 논문은 외판원 문제의 해를 쉽게 구하는 알고리즘을 제안하였다. 사전에, n(n-1)개의 데이터에 대해 각 정점에서의 거리 오름차순으로 정렬시켜 최단거리 상위 10개인 10n개를 결정하였다. 첫 번째로, 각 정점 $v_i$의 최단거리인 $r_1=d\{v_i,v_j\}$로 연결된 부분경로를 하나의 지역으로 결정하였다. $r_2$에 대해서는 지역 내 정점간 간선은 무조건 연결하고, 지역간 간선은 연결 규칙을 적용하였다. 전체적으로 하나의 해밀턴 사이클이 형성될 때까지 $r_3$ 부터는 지역간 간선만 연결하는 방법으로 정복하였다. 따라서 제안된 방법은 지역분할정복 방법이라 할 수 있다. 실제 지도상의 도시들인 TSP-1(n=26) TSP-2(n=42)와 유클리드 평면상에 랜덤하게 생성된 TSP-3(n=50)에 대해 제안된 알고리즘을 적용한 결과 TSP-1과 TSP-2는 최적해를 구하였다. TSP-3에 대해서는 Valenzuela와 Jones의 결과보다 거리를 단축시킬 수 있었다. 전수탐색 방법은 n!인데 반해, 제안된 알고리즘의 수행복잡도는 $O(n^2)$이며, 수행횟수는 최대 10n이다.

This paper introduces a 'divide-and-conquer' algorithm to the travelling salesman problem (TSP). Top 10n are selected beforehand from a pool of n(n-1) data which are sorted in the ascending order of each vertex's distance. The proposed algorithm then firstly selects partial paths that are interconnected with the shortest distance $r_1=d\{v_i,v_j\}$ of each vertex $v_i$ and assigns them as individual regions. For $r_2$, it connects all inter-vertex edges within the region and inter-region edges are connected in accordance with the connection rule. Finally for $r_3$, it connects only inter-region edges until one whole Hamiltonian cycle is constructed. When tested on TSP-1(n=26) and TSP-2(n=42) of real cities and on a randomly constructed TSP-3(n=50) of the Euclidean plane, the algorithm has obtained optimal solutions for the first two and an improved one from that of Valenzuela and Jones for the third. In contrast to the brute-force search algorithm which runs in n!, the proposed algorithm runs at most 10n times, with the time complexity of $O(n^2)$.

키워드

참고문헌

  1. A. Likas and V. T. Paschos, "A Note on a New Greedy-solution Representation and a New Greedy Parallelizable Heuristic for the Traveling Salesman Problem," Chaos, Solutions and Fractals, Vol. 13, pp. 71-78, 2002, doi:10.1016/S0960-0779(00)00227-7.
  2. A. Schrijver, "On the History of Combinatorial Optimization (till 1960)," in Handbook of Discrete Optimization'' (K. Aardal, G.L. Nemhauser, R. Weismantel, eds.), Elsevier, Amsterdam, pp. 1-68, http://homepages.cwi.nl/-lex/files/histco.pdf, 2005.
  3. J. Denzinger, D. Fuchs, M. Fuchs, and M. Kronenburg, "The Teamwork Method for Knowledge-Based Distributed Search: The travelling salesman problem," University of Kaiserslautern, 2008.
  4. S. Vempala, "18.433 Combinatorial Optimization: NP-completeness," http://ocw.mit.edu/NR/rdonlyres/Mathematics/18-433Fall2003/778D00DB-F21C-486C-ABD8-F5E7F5C929C3/O/I20.pdf, 2003.
  5. L. Stougie, "2P350: Optimaliseringsmethoden," http://www.win.tue.nl/-leen/OW/2P350/Week8/week8.pdf, College Wordt ggeven op vinjdagmiddag, 2001.
  6. M. Lihoreau, L. Chittka, and N. E. Raine, "Travel Optimization by Foraging Bumblebees Through Readjustments of Traplines after Discovery of New Feeding Locations," The American Naturalist, Vol. 176, No. 6, pp. 744-757, 2010, doi:10.1086/657042.
  7. E. Charniak and M. Herlihy, "CSC 751 Computational Complexity: Local Search Heuristics," Department of Computer Science, Brown University, 2008.
  8. D. S. Johnson and L. A. McGeoch, "The Traveling Salesman Problem and Its Variations," Kluwer Academic Publishers, pp. 369-443, 2002.
  9. A. Kazakov, R. A. Watson, and M. Zwolinski, "Traveling Salesman Problem: Local Search and Divide and Conquer Working Together," Independent Research Review, pp. 1-8, University of Southmapton, 2009.
  10. C. L. Valenzuela and A. J. Jones, "Evolutionary Divide and Conquer (I): A Novel Genetic Approach to the TSP," Evolutionary Computation, Vol. 1, No. 4, pp. 313-333, 1994, doi:10.1162/evco.1993.1.4.313.
  11. F. Bock, "An Algorithm for Solving Traveling-Salesman and Related Network Optimization Problems," The 14th ORSA National Meeting, 1958.
  12. M. M. Flood, "The Traveling-Salesman Problem," Operations Research, Vol. 4, No. 1, pp. 61-75, 1956, doi:10.1287/opre.4.1.61.
  13. G. A. Croes, "A Method for Solving Traveling Salesman Problems," Operations Research, Vol. 6, pp. 791-812, 1958, doi:10.1287/opre.6.6.791.
  14. S. Lin, "Computer Solutions of the Traveling Salesman Problem," Bell System Technical Journal, Vol. 44, pp. 2245-2269, 1965, doi:10.1002/j.1538-7305.1965.tb04146.
  15. L. H. Chuin, "IS 703: Decision Support and Optimization," School of Information Systems," Department of Computer Science, Brown University, 2008.
  16. J. Pleines, "ZIP-Methode: ein Kombinatorischer Ansatz zur Optimalen Losung Allgemeiner Traveling-Salesman-Problem (TSP)," Konnen bekannte Losungen nicht nur auf Gesamtgrphen sondern auf Teilgraphen angewandt werden, so bringt die ZIP-Methode den entscheidenden Quantensprung der rechentechnischen Vereinfachung, http://www.jochen-pleines.de/download /ZIP2006.pdf, 2006.
  17. G. Dantzig, R. Fulkerson, and S. Johnson, "Solution of a Large-scale Traveling-Salesman Problem," The Rand Corporation, http://www.cse.wustl.edu/-chen/7102/TSP.pdf, 1954.