An Initialization Model with Reference to Finite Number of Failures
|
|
|||||||||||
Détails bibliographiques
| Auteur(s): |
Atsukuni Kajima
(West Nippon Expressway Engineering Chugoku Co., Ltd., Hiroshima, Japan)
Kiyoshi Kobayashi (Kyoto University, Graduate School of Management, Kyoto, Japan) Kengo Obama (Osaka University, Graduate School of Engineering, Osaka, Japan) Kiyoyuki Kaito (Osaka University, Graduate School of Engineering, Osaka, Japan) |
||||
|---|---|---|---|---|---|
| Médium: | papier de conférence | ||||
| Langue(s): | anglais | ||||
| Conférence: | IABSE Symposium: Tomorrow’s Megastructures, Nantes, France, 19-21 September 2018 | ||||
| Publié dans: | IABSE Symposium Nantes 2018 | ||||
|
|||||
| Page(s): | S5-165 | ||||
| Nombre total de pages (du PDF): | 8 | ||||
| DOI: | 10.2749/nantes.2018.s5-165 | ||||
| Abstrait: |
In order to manage the expressway assets, it is important to properly predict the scale of failures occurring in infrastructures in future. In this paper, the authors propose an initialization time model to predict the scale of failures. Specifically, the number of latent failure places where there are possibility of failure occurrence is assumed to be followed to Poisson distribution. Furthermore, a time-dependent probability of a failure occurrence in latent failure places is expressed by using Weibull deterioration hazard model. In consequence, an initialization time model is formulated as a mixed probability model. Lastly, the authors discuss the usefulness of the model in this study through the empirical analysis of actual expressway assets. |
||||
| Mots-clé: |
pont
|
||||