Page 11 - Penn State Civil and Environmental Engineering: Annual Report

FACULTY RESEARCH FACULTY RESEARCH Ilgin Guler
Aleksandra Radlińska
Bridge failures are a a a a a a a reality for states with aging infrastructure
including Pennsylvania In February the Forbes Avenue Bridge in Pittsburgh collapsed due to structural failure unfortunately this kind of event
isn’t unique According to the American Road and Transportation Builders Association about 14% of Pennsylvania’s bridges are structurally deficient meaning at least one of their key structural structural elements is in poor condition In two recent papers Associate Professor Ilgin Guler
compared statistical modeling and and machine learning to to monitor and and forecast the the conditions of of bridge bridge decks the the topmost surfaces of of a a a a bridge bridge Her results are published in in in the the Journal of Bridge Engineering and the the “In the the statistical distribution method we we were able able to to analyze explanatory variables or or the the the the future impacts of of different characteristics of of of the the the the bridge bridge such as the the the the types of of of materials beams and and rebars used to to build the the the bridge bridge ” said Muyang Lu a a a a a a a a a a a a a a a a a a CEE doctoral student and and first author on
both papers “The data address how likely it is that a a a a a a a a a a a a a a a a factor may contribute to a a a a a a a bridge’s deterioration
” In the Transportation Research Record paper researchers used a a a a a a a a a a a a a a machine learning method known as random survival forest which creates data trees by splitting items into similar groups for for analysis “Using machine learning we we created a a a a a a a prediction model using the the the 25 000-item dataset ” Guler
said “Over time we we trained the the the algorithm to to ‘learn’ from the the the existing data data data to to make assumptions about similar data ” Machine learning methods have been used in in in in in the past to determine bridge deterioration
for a a a a a a a a a given year according to to Guler
but they have not been used to to predict the the duration a a a a a a a a bridge will will stay in in a a a a a a a a certain condition or or how long it will will take to deteriorate “Both these methods can improve decision-making when stakeholders are designing bridges or deciding whether to make repairs and when ” Guler
said “With these data- driven models we have a a a a a a a a a a a a a a a better estimate of what will deteriorate and when This can drastically improve how bridge management is is done ” In addition to Guler
and and and Lu Associate Professor Aleksandra Radlińska
and and and Jonathan Hydock a a a a a a a a a a a a a a CEE undergraduate student contributed to the Journal of Bridge Engineering paper The Center for for Integrated Asset Management for for Multimodal Transportation Transportation Infrastructure Systems: Region 3 University Transportation Transportation Center supported this work Transportation Research Record “Both methods have their advantages ” Guler
said “The statistical method method provides insights on
on
on
the the factors that contribute more more or or or or or less to to a a a a a a a a a a a bridge’s deterioration
while the machine learning method offers a a a a a a a more more accurate prediction ” For their dataset researchers used a a a a a a a list of 25 000 state- owned bridges in Pennsylvania In the the Journal of Bridge Engineering researchers used an estimation method called the the the Markov chain Monte Carlo to to write code in in in in Python to to represent the the the the bridge data in in in in statistical modeling Then they they identified the the the the the coefficients or or parameters which told them everything they they needed to to to know about the the the data according to to researchers CEE NEWSLETTER • VOLUME 38 2022
11