Stochastic Modeling of Driver Decision at the Onset of a Yellow Indication at Signalized Intersections

摘要

This paper introduces two logistic statistical models for the driver stop-run decision at the onsetof yellow at signalized intersections to capture the stochastic nature of the driver stop-rundecision. One model is a classical frequentist model, while the other uses a Bayesian statisticsapproach. The Bayesian model parameters are calibrated using the Markov Chain Monte Carlo(MCMC) slice procedure implemented within the MATLAB software. Both models aredeveloped using 3,328 stop-run records that were collected in a field experiment on the VirginiaSmart Road. The variables included in the model are driver gender, age, time-to-intersection,yellow time, approaching speed and speed limit. Both models are shown to be consistent. Inaddition, two procedures for the Bayesian model application are illustrated; namely Cascadedregression and Cholesky decomposition. Both procedures are demonstrated to producereplications that are consistent with the Bayesian model realizations, while these procedurescapture the parameter correlations without the need to store the set of parameters realizations.The Bayesian model is found to produce valid and transferable behavior by replicating multipleexperimental results. The proposed Bayesian approach is ideal for modeling multi-agent systemsin which each agent has its own unique set of parameters.