Application of Data Science and Big Data Analytics in Underground Transportation Infrastructure

REU: Underground Infrastructure REU | AUTHOR: Juliet Malkowski – California State University Los Angeles

MENTOR: Mohammad Pourhomayoun – California State University Los Angeles

ABSTRACT

Artificial Intelligence (AI) is a rapidly growing field of research that has the capability to accurately analyze and predict various outcomes in datasets once the program is given a reliable dataset and trained. There is currently an influx of data being collected faster and in larger quantities than is possible to be analyzed by a team of engineers. In the case of underground transportation infrastructure, common uses for AI include predicting tunnel boring conditions, analyzing cracks in surfaces, and predicting concrete strength. AI can be used to collect and analyze a given dataset using different machine learning methods to predict the most optimal solutions for any problem.
In my research project, I focused on predicting concrete strength under different mixing conditions. The conditions that I focused on were the amount of concrete, blast furnace slag, fly ash, amount of water, superplasticizer, coarse aggregate, and fine aggregate added to the mixture as well as how old the concrete was. Two machine learning models were used to predict concrete strength. The first model was a random forest regressor model where the accuracy of my model was analyzed using root mean square error. The second model was a linear regression model where different features were plotted against each other to determine which features had the highest impact on concrete strength and to provide a visualization for what concrete strength is expected under different feature conditions. Using these two machine learning models, expected concrete strength can be predicted.

AUTHOR BIOGRAPHY

Juliet Malkowski is a rising senior at the University of Vermont majoring in environmental engineering. Her research this summer at the California State University Los Angeles has focused on learning machine learning in the computer science department.