Modeling of the Drivers’ Decision-Making Behavior During Yellow Phase

Abstract

Stop and go modeling at signalized intersections under heterogeneous traffic conditions has remained one of the most sought-after fields. Drivers are often caught up in the dilemma zone and unable to take quick decisions whether to stop or cross the intersection. This hampers the traffic movement and may lead to accidents. These variables include distance-to-stop line, time-to-stop line, approach speed, acceleration/deceleration and category of the vehicle. Further, using external validation, the overall accuracy levels of the drivers’ decision models based on Logistic regression (81.5 percent), Fuzzy Logic (86.98 percent) and ANN (88.67 percent) were compared. Further, a hybrid surrogate model, incorporating the ‘Weighted Average’ technique was developed so that the individual disparities were diminished and the overall accuracy for both stopping and crossing vehicles was improved. This Weighted Average Hybrid Model (WAHM), formulated by coupling Fuzzy Logic and ANN, yielded a highly accurate result (96.15 percent) and outperformed both ANN and Fuzzy Logic.