Citation

Weissmann J, Weissmann AJ, Montoya A. Transp. Res. Rec. 2023; 2677(7): 307-316.

Copyright

(Copyright © 2023, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

10.1177/03611981231152460

PMID

unavailable

Abstract

This paper discusses 38 bridge- and eight culvert deterioration models of three bridge and one culvert condition rating stored in the National Bridge Inventory (NBI) as Items 58, 59, 60 and 62. The models were developed by age groups and by families of relevant variables such as span type, deck material, and bridges over water or land. Each model is a 2-year Markov matrix for every age group or family, containing the probabilities that each rating will remain the same, or transition to all possible lower ratings. These probabilities were calculated by counting all actual transitions in a database containing 19 years of bridge inspection data in Texas, unlike previous models which assumed that ratings cannot drop by more than 1 point in one inspection cycle. Models were validated and standard errors calculated. Updatable results were implemented in Excel workbooks and included 18-year forecasts of condition rating and network deterioration tables and curves, and comparisons between the current network condition and 10-year forecasts. The culvert models also included updatable cost forecasts to maintain the culvert network above Condition Rating 4. The main technical contribution of this project is the probabilities of bridge and culvert ratings decreasing by more than 1 in a 2-year inspection cycle being greater than zero for all condition ratings modeled (range: 8% to 61%). The models developed using the Texas data were strictly applicable to maintenance, environmental, and traffic conditions comparable to those prevalent in Texas. However, the modeling methodology is applicable to any state or location.

Language: en