Deep Learning Approach for Predictive Analytics to Support Diversion during Freeway Incidents

Abstract

Route diversion during incidents on freeways has been proven to be a useful tactic to mitigate non-recurrent congestion. However, the capacity constraints created by the signals on the alternative routes put limits on the diversion process since the typical time-of-day (TOD) signal control cannot handle the sudden increase in the traffic on the arterials because of diversion. Thus, there is a need for active transportation management strategies that support agencies in identifying the potential diversion routes for freeway incidents and the need for adjusting the traffic signal timing under different incident and traffic conditions. This paper investigates the use of a data analytic approach based on the long short-term memory (LSTM) deep neural network method to predict the alternative routes dynamically using incident attributes and traffic status on the freeway, and travel time on both the freeway and alternative routes during the incident. Additionally, a methodology is proposed for the development of special signal plans for the critical intersections on the alternative arterials based on the results from the LSTM neural network, combined with simulation modeling, and signal timing optimization. The methodology developed in the paper can be easily implemented by the transportation agencies, as it is based on data that are generally available to the agencies. The results from this paper indicate that the developed methodology can be used as part of a decision support system (DSS) to manage the traffic proactively during the incidents on the freeways.

Data and Modeling Support of the Management of Diversion Routes During Freeway Incidents

“…In a previous study, the authors developed a machine learning-based method to identify the critical alternative routes that motorists used for the diversion during incidents. The details of the study are discussed elsewhere ( 4 ). Figure 6 shows the alternative route identified for the network used as a case study in this paper.…”

“…The alternative route includes a north–south arterial (S Military Trail) parallel to the freeway (I-95) connected through four small east–west arterial segments. During incidents, traffic can use these links to divert based on the traffic and incident attributes, as discussed in Saha et al ( 4 ). An intersection (Palmetto Park Rd-S Military Trail) was selected for demonstration of the activation of special signal plan generation based on the methodology developed in this study.…”

“…The diversion rate is expected to be constrained by the capacities of the freeway off-ramps and the signals that are close to these ramps. Thus, there is a need for cost-effective methods to predict diversion considering the above factors ( 4 , 5 ).…”

Combining Machine Learning and Fuzzy Rule-Based System in Automating Signal Timing Experts’ Decisions during Non-Recurrent Congestion

“…In those two scenarios, signal timing plans along the arterial were modified to increase the capacity of the main approaches by increasing the cycle length and the relative green time for the main direction while the green time for the side streets was kept constant. Saha et al ( 12 ) developed methods for the selection of special signal timing plans to accommodate traffic diversion during freeway incidents to arterial streets.…”

Large-Scale Road Network Congestion Pattern Analysis and Prediction Using Deep Convolutional Autoencoder

“…); it shows the likelihood of traffic jam occurrence on each road in a transportation network by monitoring the patterns from the historical data. The proper monitoring of the congestion pattern is the first step toward building a resilient and sustainable transportation system [13][14][15][16][17]. Based on the congestion pattern map, the traffic planning agencies, policy-makers, and transportation experts can pinpoint and investigate those networks with a high probability of being in traffic-jam states, examine the cause and, accordingly, plan a demand-side management-based policy toward mitigating the congestion.…”