A critical function of intelligent transportation systems is studying and analyzing the effects of road condition variables (e.g. construction, severe weather, and the like) on traffic to aid in improving road designs, estimating travel time, and increasing safety. In this thesis, Multivariate Distance Matrix Regression (MDMR), a well-studied algorithm applied in brain research, is explored and applied in the transportation domain to assess the relationship and the effects of traffic conditions on transportation system performance. The Multivariate Distance Matrix Regression (MDMR) is utilized to study the relationship between input experimental factors and the association of response variables. When studying transportation, input factors can be represented as any factor that may have an effect on traffic, and response variables can be represented by traffic speed values over time for each segment of a road. The output is represented as a probability Value (P-Value) for each segment of the road as an indication of an effect of the studied factor on that specific segment. The National Performance Management Research Dataset (NPMRDS), (i.e., a probe-based traffic dataset) was used to study traffic performance based on specific factors by applying MDMR under different traffic scenarios.Moreover, a novel clustering algorithm for time series data is proposed by optimizing the F-statistic (i.e., a measurement metric to study the significance difference of two or more groups) to find the best segregation of time series between two or more groups. The clustering algorithm gave promising preliminary results when compared with K-means.