Many aspects of transportation infrastructure have benefited from the use of full-scale management systems to quantify condition and predict useful life estimates. Although culvert management models have been tested in the past, none has proven functional. In this study, a condition rating system for culvert repair or replacement was developed. Parameters that describe the condition of an existing culvert were tested for potential as predictors of overall culvert condition. These 33 parameters were recorded at 460 culverts distributed geographically throughout Montana. An ordered probit statistical model indicated that 9 of the initial 33 potential predictors were statistically significant. Measurements of these nine parameters can be used in the resulting model to classify a culvert into a 1-to-5 condition ranking, in which 5 is the best condition and 1 is the worst.
Samenvatting