DocumentCode :
1826547
Title :
Generating large scale undirected graph for solving flow network problems
Author :
Chen, S.G.
Author_Institution :
Dept. of Ind. Eng. & Manage., Tungnan Univ., Taipei, Taiwan
fYear :
2010
fDate :
7-10 Dec. 2010
Firstpage :
1334
Lastpage :
1338
Abstract :
Since network analysis has been a formal topic in Operations Research as well as in Reliability, researchers may need to validate their models or theories with large scale feasible graphs for test. This paper proposes an algorithm to generate a large scale undirected capacitated/uncapacitated graph for applications. Conventionally, NETGEN is employed for such purposes. But this program is about 30 years old and no algorithm available for verifying the properties of the generated graphs. Moreover, it only created directed graph. For undirected flow network problems, there is no algorithm available in the literature. In this paper, a theoretical aspect and new features are addressed. This algorithm is not an extension of NETGEN. The undirected versions of the 40 benchmarks in NETGEN are inspected in this paper. NETGEN had a limited size of 8000 nodes and 35000 arcs in different graphs. The limits of nodes and arcs are not constrained in the proposed algorithm. Unlike NETGEN involving thousand lines of codes, a sample code of no more than 43 lines is tested for examples.
Keywords :
graph theory; NETGEN; flow network problems; undirected capacitated graph; undirected uncapacitated graph; Artificial neural networks; Benchmark testing; Computer network reliability; Operations research; Reliability engineering; Telecommunication network reliability; NETGEN; Undirected Graph; capacitate; maximal flow; minimal cost;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Industrial Engineering and Engineering Management (IEEM), 2010 IEEE International Conference on
Conference_Location :
Macao
ISSN :
2157-3611
Print_ISBN :
978-1-4244-8501-7
Electronic_ISBN :
2157-3611
Type :
conf
DOI :
10.1109/IEEM.2010.5674412
Filename :
5674412
Link To Document :
بازگشت