Analyzing Taiwan Bridge Management System for Decision Making in Bridge Maintenance - A Big Data Approach

Abstract

The Taiwan Bridge Management System (TBMS) has been online for 15 years and has an inventory of 33,275 bridges, including all kinds of bridges and culverts in Taiwan. Currently, the number of fields in all tables in the databases of TBMS is around 6,500 with more than 3 million data records in its databases. Meanwhile, bridge inspection data and maintenance data are increasing at a speed of 15,000 records annually. Thus, the TBMS databases are deemed as “Big Data.” There are more than 9,500 bridges that are over 20 years old with another 7,200 bridge having unknown built years in the TBMS. The bridges in Taiwan have stepped into the stage where maintenance is crucial and frequently required. Therefore, this research aims at analysing the database in the TBMS using “Big Data” approach for determining maintenance strategies for these bridges. This paper describes results of the first year's research efforts. Relevant literature in bridge maintenance, prioritization, and life-cycle bridge management were firstly reviewed. Concepts, theories, techniques, and available software for analysing “Big Data” were also intensively examined and summarized. In next year, functions will be programmed and applied to the TBMS databases using appropriate “Big Data” software to obtain useful information in bridge deterioration, repair methods, and maintenance costs.