Home // International Journal On Advances in Intelligent Systems, volume 13, numbers 1 and 2, 2020 // View article
Authors:
Prasad Chetti
Hesham Ali
Keywords: Structural Health Monitoring; Population Analysis; Correlation Networks; Markov Clustering; Sufficiency Rating; National Bridge Inventory database
Abstract:
Many recent studies have shown that a large percentage of bridges in many parts of the world have a low safety rating. The national bridge inventory database contains data on more than 600,000 bridges, where each bridge has 116 parameters. Current safety inspections require bridge inspectors to manually inspect each bridge every few years. Manpower and budget constraints limit such inspections from being performed more frequently. More efficient approaches need to be developed to improve the process of bridge inspection and increase the overall safety of bridges and civil infrastructures. In this study, we propose a correlation network model to analyze and visualize the big data associated with more than 600,000 bridges in the national bridge inventory database. We use correlation networks based on various safety parameters, then apply the Markov clustering algorithm to analyze a sub-set of 9,546 steel-stringer/multi-beam or girder bridges. We use the produced clusters to propose a different maintenance schedule based on the bridges that show a higher chance of becoming deficient. Results show that of the top five clusters of bridges, three need to be serviced more frequently. We recommend that their inspection frequency be reduced to 12 months instead of 24 months.
Pages: 151 to 162
Copyright: Copyright (c) to authors, 2020. Used with permission.
Publication date: June 30, 2020
Published in: journal
ISSN: 1942-2679