### Abstract

The performance of Monte-Carlo Simulation(MCS) is highly related to the number of simulation. This paper introduces a hypothesis testing technique and incorporated into a Particle Swarm Optimization(PSO) based Monte-Carlo Simulation(MCS) algorithm to solve the complex network reliability problem. The function of hypothesis testing technique is to reduce the dispensable simulation in network system reliability estimation. The proposed technique contains three components: hypothesis testing, network reliability calculation and PSO algorithm for finding solutions. The function of hypothesis testing is to abandon unpromising solutions; we use monte-carlo simulation to obtain network reliability; since the network reliability problem is NP-hard, PSO algorithm is applied. Since the execution time can be better decreased with the decrease of Confidence level of hypothesis testing in a range, but the solution becomes worse when the confidence level exceed a critical value, the experiment are carried out on different confidence levels for finding the critical value. The experimental results show that the proposed method can reduce the computational cost without any loss of its performance under a certain confidence level.

Language | English |
---|---|

Title of host publication | Proceedings of the 8th International Conference on Advanced Computational Intelligence, ICACI 2016 |

Publisher | Institute of Electrical and Electronics Engineers Inc. |

Pages | 310-317 |

Number of pages | 8 |

DOIs | |

Publication status | Published - 7 Apr 2016 |

Event | 8th International Conference on Advanced Computational Intelligence, ICACI 2016 - Chiang Mai, Thailand Duration: 14 Feb 2016 → 16 Feb 2016 |

### Conference

Conference | 8th International Conference on Advanced Computational Intelligence, ICACI 2016 |
---|---|

Country | Thailand |

City | Chiang Mai |

Period | 14/02/16 → 16/02/16 |

### Fingerprint

### Keywords

- hypothesis testing
- Monte-Carlo simulation
- network reliability
- network reliability optimization
- particle swarm optimization

### Cite this

*Proceedings of the 8th International Conference on Advanced Computational Intelligence, ICACI 2016*(pp. 310-317). [7449844] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICACI.2016.7449844

}

*Proceedings of the 8th International Conference on Advanced Computational Intelligence, ICACI 2016.*, 7449844, Institute of Electrical and Electronics Engineers Inc., pp. 310-317, 8th International Conference on Advanced Computational Intelligence, ICACI 2016, Chiang Mai, Thailand, 14/02/16. https://doi.org/10.1109/ICACI.2016.7449844

**Particle swarm optimization with Monte-Carlo simulation and hypothesis testing for network reliability problem.** / Wu, Lu Yao; Chen, Wei Neng; Deng, Hao Hui; Zhang, Jun; Li, Yun.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

TY - GEN

T1 - Particle swarm optimization with Monte-Carlo simulation and hypothesis testing for network reliability problem

AU - Wu, Lu Yao

AU - Chen, Wei Neng

AU - Deng, Hao Hui

AU - Zhang, Jun

AU - Li, Yun

PY - 2016/4/7

Y1 - 2016/4/7

N2 - The performance of Monte-Carlo Simulation(MCS) is highly related to the number of simulation. This paper introduces a hypothesis testing technique and incorporated into a Particle Swarm Optimization(PSO) based Monte-Carlo Simulation(MCS) algorithm to solve the complex network reliability problem. The function of hypothesis testing technique is to reduce the dispensable simulation in network system reliability estimation. The proposed technique contains three components: hypothesis testing, network reliability calculation and PSO algorithm for finding solutions. The function of hypothesis testing is to abandon unpromising solutions; we use monte-carlo simulation to obtain network reliability; since the network reliability problem is NP-hard, PSO algorithm is applied. Since the execution time can be better decreased with the decrease of Confidence level of hypothesis testing in a range, but the solution becomes worse when the confidence level exceed a critical value, the experiment are carried out on different confidence levels for finding the critical value. The experimental results show that the proposed method can reduce the computational cost without any loss of its performance under a certain confidence level.

AB - The performance of Monte-Carlo Simulation(MCS) is highly related to the number of simulation. This paper introduces a hypothesis testing technique and incorporated into a Particle Swarm Optimization(PSO) based Monte-Carlo Simulation(MCS) algorithm to solve the complex network reliability problem. The function of hypothesis testing technique is to reduce the dispensable simulation in network system reliability estimation. The proposed technique contains three components: hypothesis testing, network reliability calculation and PSO algorithm for finding solutions. The function of hypothesis testing is to abandon unpromising solutions; we use monte-carlo simulation to obtain network reliability; since the network reliability problem is NP-hard, PSO algorithm is applied. Since the execution time can be better decreased with the decrease of Confidence level of hypothesis testing in a range, but the solution becomes worse when the confidence level exceed a critical value, the experiment are carried out on different confidence levels for finding the critical value. The experimental results show that the proposed method can reduce the computational cost without any loss of its performance under a certain confidence level.

KW - hypothesis testing

KW - Monte-Carlo simulation

KW - network reliability

KW - network reliability optimization

KW - particle swarm optimization

UR - http://www.scopus.com/inward/record.url?scp=84966593325&partnerID=8YFLogxK

U2 - 10.1109/ICACI.2016.7449844

DO - 10.1109/ICACI.2016.7449844

M3 - Conference contribution book

SP - 310

EP - 317

BT - Proceedings of the 8th International Conference on Advanced Computational Intelligence, ICACI 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -