Machine Learning Approach to Model Interdependent Network Performance

Ramineni Chittibabu Naidu, Ghaneshvar

Machine Learning Approach to Model Interdependent Network Performance

dc.contributor.advisor	Barker, Kash
dc.contributor.author	Ramineni Chittibabu Naidu, Ghaneshvar
dc.contributor.committeeMember	Radhakrishnan, Sridhar
dc.contributor.committeeMember	González, Andrés
dc.date.accessioned	2020-01-06T17:01:00Z
dc.date.available	2020-01-06T17:01:00Z
dc.date.issued	2019-12-13
dc.date.manuscript	2019-12-10
dc.description.abstract	Resilience can be understood as the property of an object or system to recover from its state of disruption back to its complete functional stage as it was prior to the disruption event. There are different ways of measuring the resilience of a system and tracking the system performance is one of the methods. Thus, measuring the time taken by the system to recover to its original state is one of the parameters that can be considered. In this research work, we have focused on building a model(s) that predicts the time taken for an interdependent network to recover and function at 100%. In order to implement this idea, we present a case-study of the system of interdependent water, gas, and power utilities in Shelby County, TN. The model is trained using the train data set from the data set generated by running the optimization code multiple times and observing the time taken for the inter-dependent network to recover completely. The prediction of time to recover is made on the test data set using different models and the results are then compared.	en_US
dc.identifier.uri	https://hdl.handle.net/11244/323266
dc.language	en	en_US
dc.subject	Resilience.	en_US
dc.subject	Predictive Analysis.	en_US
dc.subject	Machine Learning.	en_US
dc.subject	Network Performance.	en_US
dc.thesis.degree	Master of Science	en_US
dc.title	Machine Learning Approach to Model Interdependent Network Performance	en_US
ou.group	Gallogly College of Engineering	en_US
shareok.nativefileaccess	restricted	en_US