Date of Award:


Document Type:


Degree Name:

Master of Science (MS)


Computer Science


Douglas Galarus


Vladimir Kulyukin

Third Advisor:

Nicholas Flann


In the states which record extreme weather conditions and high snow in winters, the travel time to drive between cities can get highly affected due to these bad weather conditions. The present solutions to tackle this problem are largely flow or time related and do not take weather conditions into account while making the predictions about travel time. Also these solutions can mostly be used for real time travel and not the future travel. In addition to that, the studies that have been done in this space are mostly for urban travel times but most parts of the interstate highways go through rural areas.

This thesis is an attempt to analyze the impact of weather conditions on the travel time or speed of vehicles on highways during winter months. It uses the weather forecast data to make predictions on the speeds of various highway segments. The problem of predicting the speed has been treated as a regression problem and various statistical measures and machine learning techniques have been used to gain some insights on the behavior of speeds for various road segments on highways.