1995 Fiscal Year Final Research Report Summary
Study on largescale combinatorial optimization problems by graph-network
| Project/Area Number |
05452210
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
System engineering
|
|---|
| Research Institution | KYOTO UNIVERSITY |
|---|
Principal Investigator |
IBARAKI Toshihide Kyoto University, Graduate School of Engineering, Professor, 工学研究科, 教授 (50026192)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
MASUYAMA S. Toyohashi Inst.of Technology, Faculty of Engineering, Assoc.Prof., 工学部, 助教授 (60173762)
YAGIURA M. Kyoto Univ.Graduate School of Eng., Assistant Prof., 工学研究科, 助手 (10263120)
IBARAKI Satoru Kyoto Univ.Graduate School of Eng., Assistant Prof., 工学研究科, 助手 (10252488)
NAGAMOCHI Hiroshi Kyoto Univ.Graduate School of Engineering, Assoc.Prof., 工学研究科, 助教授 (70202231)
|
|---|
| Project Period (FY) |
1993 – 1995
|
| Keywords | graphs / networks / combinatorial optimization / metaheuristics / genetic algorithms |
|---|
| Research Abstract |
As evidenced by the thoretical results of NP-hardness, it is well known that most of the combinatorial optimization problems are intractable. To overcome this difficulty, it is necessary to utilize the inherent structures of the problems at hand. A good example of this kind is the implicit network structures, which are often found in many problems in practice.
This study pursues this direction in two respects, that is, to find efficient algorithms for some of the graph problems, and to make use of such algorithms to enhance the algorithms developed for solving various combinatorial optimization problems.
As an example of the first possibility, we proposed a new efficient algorithm to compute minimum cuts in a graph, and implemented it to examine its computational performance. We furthermore used this algorithm as subalgorithms of various graph problems such as the optimum graph augmentation problem.
We also considered metaheuristics, which include such algorithms as genetic algorithms, tabu search and simulated annealing, as practical tools to solve combinatorial optimization problems approximately. To compare their performance, we conducted extensive computational experiment on a test-bed of the single machine scheduling problem.
|
