Repository logo
  • English
  • Català
  • Čeština
  • Deutsch
  • Español
  • Français
  • Gàidhlig
  • Italiano
  • Latviešu
  • Magyar
  • Nederlands
  • Polski
  • Português
  • Português do Brasil
  • Suomi
  • Svenska
  • Türkçe
  • Tiếng Việt
  • Қазақ
  • বাংলা
  • हिंदी
  • Ελληνικά
  • Yкраї́нська
  • Log In
    New user? Click here to register. Have you forgotten your password?
Repository logo
  • Communities & Collections
  • All of DSpace
  • English
  • Català
  • Čeština
  • Deutsch
  • Español
  • Français
  • Gàidhlig
  • Italiano
  • Latviešu
  • Magyar
  • Nederlands
  • Polski
  • Português
  • Português do Brasil
  • Suomi
  • Svenska
  • Türkçe
  • Tiếng Việt
  • Қазақ
  • বাংলা
  • हिंदी
  • Ελληνικά
  • Yкраї́нська
  • Log In
    New user? Click here to register. Have you forgotten your password?
  1. Home
  2. Browse by Author

Browsing by Author "Reyes Pedernera, Lorenzo"

Now showing 1 - 1 of 1
Results Per Page
Sort Options
  • Loading...
    Thumbnail Image
    Item
    Identificación de variables influyentes en calificación de condición de deterioro de puentes usando algoritmos de Machine Learning.
    (Universidad de Concepción, 2024) Reyes Pedernera, Lorenzo; Echaveguren Navarro, Tomás Benjamín
    Chile has detailed databases of traffic and bridge inventory and inspection that are not connected. The bridge inspection includes the bridge condition rating, used to make maintenance decisions but it cannot predict the bridge condition in time. To calibrate a bridge condition behavior model, it is crucial to determine which explanatory variables need to be considered, as including or excluding significant explanatory variables or weighting variables wrongfully may affect the prediction ability of the behavior model. The purpose of this research was to determine the influence of explanatory variables in the bridge condition rating using Chilean databases. The influence of variables was measured using ReliefF and Feature Importance by Permutation joined with Random Forest and Extreme Gradient Boosting. In conclusion the most influential variables weighted by percentage score were the total number of deteriorations in the bridge, total bridge length, total bridge width, number of deteriorations by bridge longitude and girder material. Among the influential variables only the total number of deteriorations and the cause of deterioration are included in the bridge condition rating calculations. These variables can inform the calibration of bridge condition behavior models.
Síguenos en...