## Abstract

In this paper, the project earthquake occurrence risk coefficient is determined for each construction project that is located in one of Iran's twenty seismic regions. This coefficient is allocated, regardless of the current situation of the project, being in the plan or execution phase or even completed. This coefficient indicates the possibility of an earthquake occurrence during a project's life time. To find this coefficient, the Gutenberg-Richter linear relationship has been applied, in conjunction with the Poisson distribution. The Gutenberg-Richter linear equation expresses the relationship between the magnitude of an earthquake and the number of occurrences, during a fixed time, of that magnitude. To find the linear relationship for a series of earthquakes with different magnitudes occurring in the same seismic region, the Ordinary Least Square (OLS) has been used. Two linear regression assumption violations, which are variance heteroscedasticity and autocorrelation, have been tested on the available data. In the case of finding one or both of these two violations, The Generalized Least Square (GLS) has been applied to produce a better regression line. Moreover, the second order type of the Gutenberg-Richter relationship has also been determined to validate the linear one. In conclusion, by application of the Poisson distribution and by having the design earthquake's magnitude and project life time, the third parameter, which is the design earthquake occurrence risk, can be determined for a given construction project in a specific location in Iran.

Original language | English |
---|---|

Pages (from-to) | 62-73 |

Number of pages | 12 |

Journal | Scientia Iranica |

Volume | 17 |

Issue number | 1 |

Publication status | Published - 2010 |

## Keywords

- Construction management
- Earthquake risk analysis
- GLS
- OLS
- Project management
- Risk management