Probabilistic traffic breakdown forecasting through Bayesian approximation using variational LSTMs

This paper proposes a framework for short-term traffic breakdown probability calculation using a Variational LSTM neural network model. Considering that traffic breakdown is a stochastic event, this forecast framework was devised to produce distributions as outputs, which cannot be achieved using standard deterministic recurrent neural networks. Therefore, the model counts on the robustness of neural networks but also includes the stochastic characteristics of highway capacity. The framework consists of three main steps: (i) build and train a probabilistic speed forecasting neural network, (ii) forecast speed distributions with the trained model using Monte Carlo (MC) dropout, and therefore perform Bayesian approximation, and (iii) establish a speed threshold that characterizes breakdown occurrence and calculate breakdown probabilities based on the speed distributions. The proposed framework produced an efficient control over traffic breakdown occurrence, can deal with many independent variables or features, and can be combined with traffic management strategies.

Automated summary

This paper proposes a framework that uses a Variational LSTM neural network model to calculate the probability of short-term traffic breakdown. Unlike standard deterministic recurrent neural networks, this framework is designed to produce output distributions, considering that traffic breakdown is a stochastic event. The framework includes the robustness of neural networks and the stochastic characteristics of highway capacity. It consists of building and training a probabilistic speed forecasting neural network, forecasting speed distributions using Monte Carlo dropout for Bayesian approximation, and establishing a speed threshold for breakdown occurrence and calculating breakdown probabilities based on the speed distributions. The proposed framework efficiently controls traffic breakdown, handles multiple independent variables or features, and can be combined with traffic management strategies.