Knowledge Systematization for Design and Experiments Utilizing Evolutionary Computation
| Project/Area Number |
15K17502
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Computational science
|
|---|
| Research Institution | Kobe University (2016)
Institute of Physical and Chemical Research (2015) |
|---|
Principal Investigator |
|
|---|
| Research Collaborator |
HATANAKA Toshiharu
Zhou Chenting
|
|---|
| Project Period (FY) |
2015-04-01 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2016: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2015: ¥2,990,000 (Direct Cost: ¥2,300,000、Indirect Cost: ¥690,000)
|
|---|
| Keywords | 実験計画 / 進化計算 / 大規模シミュレーション / 分散分析 / 重回帰分析 / 避難 / マルチエージェントシミュレーション / 防災・減災 / 社会科学 / 計算機シミュレーション / 実験計画法 / 感度分析 / 社会シミュレーション / エージェントシミュレーション |
|---|
| Outline of Final Research Achievements |
Since more number of factors like evacuation routes in a city should be treated to find significant factors, the number of simulation-executions exponentially increases. “Evolutionary Design of Experiments” is a proposed framework to reduce the number of simulation-executions.
It was reported that a benchmark problem was established. The benchmark problem has features which is based on a relationship between Tsunami evacuation-routes as simulation-inputs and Tsunami evacuation-time as simulation-output in Kanazawa region, Japan. It was also reported that significant factors could be found with less number of simulation-executions by applying a novel algorithm of Evolutionary Design of Experiments with evaluation function which evaluate sample variance of simulation-outputs to the benchmark problem.
|
Report
(3 results)
Research Products
(11 results)
