Spatio-Temporal Prediction of Freeway Congestion Patterns using Discrete Choice Methods

Predicting freeway traffic states is, so far, based on predicting speeds or traffic volumes with various methodological approaches ranging from statistical modeling to deep learning. Traffic on freeways, however, follows patterns in space-time like stop-and-go waves or mega jams. These patterns by itself are informative because they propagate in space-time in different ways, e.g., stop-and-go waves exhibit a typical that can range far ahead in time. When these patterns and their propagation become predictable, the traffic state prediction can be improved by this information. In this paper, we use a rich data set of congestion patterns on the A9 freeway in Germany close to Munich to develop a mixed logit model to predict congestion patterns. As explanatory variables, we use variables characterizing the layout of the freeway and variables describing the presence of previous congestion patterns. We find that such a mixed logit model can improve the prediction of congestion patterns compared to the prediction with the average presence of patterns at a given location. In future research, we further develop the model by integrating speed information.     «

Predicting freeway traffic states is, so far, based on predicting speeds or traffic volumes with various methodological approaches ranging from statistical modeling to deep learning. Traffic on freeways, however, follows patterns in space-time like stop-and-go waves or mega jams. These patterns by itself are informative because they propagate in space-time in different ways, e.g., stop-and-go waves exhibit a typical that can range far ahead in time. When these patterns and their propagation beco...     »