<div class="htmlview paragraph">The permanent growing traffic volume within the last years leads to novel characteristics of the occuring traffic conditions. However, stochastic prognosis systems on highway corridors still mainly rely on Markovian and ARMA processes [<span class="xref">2</span>], possessing the disadvantage of underestimating resulting peaks due to its finite memory. This fact often leads to inadequate prediction measurements and unexpected congestions. Contrary to conventional systems, fractional stochastic algorithms have proved to be a superior alternative in various fields as internet traffic [<span class="xref">5</span>], stock market prediction etc. in regard to appropriate peak modeling in the overall traffic situation as well as in microscopic particular sectors [<span class="xref">4</span>] (e.g. in order to depict platoons driving along with different speeds). Thus, this paper will first give a clear insight into the theory of fractional modeling. Special intention is laid on distinguishing different definitions of fractionality and evaluating its differences, a topic often leading to confusion in the general understanding. Furthermore two economic, adapted methods will be presented how to extend existing ARMA/Markovian systems to hybrid and/or stand-alone fractional prediction tools for highway corridors. In order to model the influence of individual ramp entries, their behavior on corridor entries can be described as heavy-distributed variables, leading to the desired peak-modeling scenario at all time scales.</div>