Binary Logistic Regression Approach for Decision Making in Bridge Management

Abstract

Bridge management professionals need effective tools to help guide the decision-making process and maintain quality infrastructure in a region. A new binary response is herein defined by categorizing bridges as at-risk and not at-risk, based on the existing overall bridge condition scores. Fitting binary logistic regression model for the response, the probability of a bridge being at-risk is expressed in terms of the primary bridge factors age, load, types of construction material and structural design, and conditions of the deck, superstructure, and substructure. These estimated probabilities multiplied by specified consequence values are used to introduce the risk classes and their ranks. Employing the method for training and validating sets of sizes 13,540 and 3,385 in 2017, and 13,481 and 3,370 in 2018 data in National Bridge Inventory (NBI) Indiana, a statistically significant model is established containing age, load, conditions of both superstructure and substructure. Moreover, at-risk bridges are identified from Indiana NBI data in both years and for a subset from Connecticut in 2017. The novel bridge-ranking tool prioritizes bridges for maintenance purposes such as replacing or repairing and hence efficiently guides the management in the decision-making process for capital expenditures, and perhaps, for predicting the missing overall bridge condition.