Enhancement of the FDOT's project level and network level bridge management analysis tools
-
2011-02-01
Details:
-
Creators:
-
Corporate Creators:
-
Subject/TRT Terms:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report; May 2008 - February 2011
-
Corporate Publisher:
-
Abstract:Over several years, the Florida Department of Transportation (FDOT) has been implementing the AASHTO Pontis Bridge Management System to support network-level and project-level decision making in the headquarters and district offices. Pontis is an integral part of a Department-wide effort to improve the quality of asset management information provided to decision makers. With the success of these previous research efforts, FDOT further investigated several additional modeling issues that were not possible during earlier Pontis implementation work.
First, a sensitivity analysis was performed on the Project Level Analysis Tool (PLAT) and Network Analysis Tool (NAT), as well as a comparison made between the PLAT and NAT models and the National Cooperative Highway Research Program (NCHRP) Report 590, which explored the criteria used for priority setting and resource allocation. The analysis suggested priority enhancements to PLAT/NAT, including improved deterioration and cost models, and multi-objective optimization. Secondly, an improved version of the NBI Translator has been developed and implemented using two years of bridge inspection data from the Florida bridge inventory. A standalone computer program was developed, as well as a Microsoft Excel spreadsheet version of the Translator program written in Visual Basic for Applications (VBA), which was incorporated into the PLAT.
Next, the research developed improved deterioration, action effectiveness, and cost models for Pontis and the PLAT. A new, simplified procedure was developed for estimating one-step Markovian models that produces usable results with significantly smaller sample sizes than traditional regression. As the fifth accomplishment, models were developed for estimating user costs at bridge sites where no detour is considered. Several existing user cost models were reviewed in the study, including some traditional roadway-based models and the previous FDOT user cost model for bridges. New accident models were formulated based on Florida crash data at bridge sites for years 2003 through 2007, including the following: binomial logistic regression, Poisson regression, and negative binomial regression models.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type: