Frequency Analysis of Bridge Condition Explanatory Data Items for Customized Data Collection and Bridge Management

摘要

Understanding the relative importance of large number of potential explanatory data items possibly explaining bridge condition deteriorations will help bridge management agencies better allocate resources for data collection, and simplify the data analysis. Previous studies explored how various factors influence bridge deteriorations and prioritized them, but they either used data from one geographic region, or used one or two statistical methods for analyzing data from the United States. Using National Bridge Inventory data, this study presents a systematic statistical investigation to better understand how the relative importance of different explanatory data items vary across regions in the U.S. and bridge condition rating items. We observed substantial variations in the frequency generated for different rating items using data from different regions, while some items are consistently identified as important. Deck material related features of bridges are items not studied in-depth in previous studies while consistently identified as important data items in this research.