Building Interpretable Methods For Identifying Bridge Maintenance Patterns

Presenter Information

Akshay KaleFollow

Advisor Information

Robin Gandhi

Location

MBSC Dodge Room 302B - G

Presentation Type

Oral Presentation

Start Date

4-3-2022 2:00 PM

End Date

4-3-2022 3:15 PM

Abstract

According to the American Road and Transportation Builder Association, approximately 47,000 or 9.1\% of bridges are structurally deficient. About 235,000 bridges or 38\% of the bridges require immediate maintenance. Bridge engineers are constantly looking for methods to extract insight from the bridge inspections records to plan bridge maintenance efficiently. Previous researchers have developed machine learning models that have identified influential factors for bridge maintenance. Despite the current understanding of significant factors that drive bridge maintenance, interactions between these influential factors that explain maintenance patterns remain incomplete. In this research study, we developed a method that adopts a decision tree model to generate a decision tree and apply associated rule mining to identify influential patterns that contribute to bridge maintenance.

This document is currently not available here.

COinS
 
Mar 4th, 2:00 PM Mar 4th, 3:15 PM

Building Interpretable Methods For Identifying Bridge Maintenance Patterns

MBSC Dodge Room 302B - G

According to the American Road and Transportation Builder Association, approximately 47,000 or 9.1\% of bridges are structurally deficient. About 235,000 bridges or 38\% of the bridges require immediate maintenance. Bridge engineers are constantly looking for methods to extract insight from the bridge inspections records to plan bridge maintenance efficiently. Previous researchers have developed machine learning models that have identified influential factors for bridge maintenance. Despite the current understanding of significant factors that drive bridge maintenance, interactions between these influential factors that explain maintenance patterns remain incomplete. In this research study, we developed a method that adopts a decision tree model to generate a decision tree and apply associated rule mining to identify influential patterns that contribute to bridge maintenance.